Electric Vehicle Battery Primer: A Conversation with John Voelcker


Batteries are often cited as the biggest hurdle to a future populated with electric vehicles. Why? We decided to ask John Voelcker — automotive editor for IEEE Spectrum, writer and consultant — for his take on the future of the automotive battery. Here’s his thoughts:

Q: John, you focus a lot on the automotive industry’s move towards cleaner energy. Can you give our readers your thoughts on the state of the art in batteries?

A: Sure. Today and for the next 10 years, in electrically driven automobiles, the state of the art is lithium-ion batteries. They’re the successor to the nickel-metal hydride (NiMH) batteries used in hybrids like the Toyota Prius. But they have roughly twice the energy density.

Remember, though, that “lithium-ion” encompasses many different chemistries. They all move lithium ions around, but the other electrode may be one—or a combination—of several different metals. And they’re very different in energy storage, power, and safety.

Q: Carmakers are certainly making great strides there. But in the technology world, mobility is also spurring innovation. Where do you think we’ll see the biggest breakthroughs—in a car, or a phone?

A: The companies that make or supply mobile phones, laptops, and other portable consumer electronics are always the most aggressive innovators. Remember, Sony invented the very first lithium-ion battery way back in 1989 for its first portable camcorder.

But cars are a very, very different kettle of fish. No one in their right mind expects their cellphone battery to work after years of steady pounding, from –30 to 120 degrees (Fahrenheit), and that it should last for TEN YEARS! But that’s what car buyers expect, so that’s what car-battery makers have to build. And that’s a pretty tall order. So the mobile phones come first, but the car batteries are the more impressive achievement.

Q: Do you think startups like Tesla will be able to survive and thrive, or do you see them being consumed by the automotive giants?

A: No one has successfully started and run a car company in the U.S. since before World War II. I think ultimately they’ll be bought. And if they’re bought for the right price, it’ll make their VCs very happy—but it seems a particularly risky way to do VC, all this long-lead, durable-goods stuff ….

Q: How efficient do you think batteries will become in five years’ time? Why?

A: The improvement in batteries’ energy density runs on a fairly steady curve, and so does the decline in cost per kWh. Think of it as a battery version of Moore’s Law. I couldn’t tell you off the top of my head what the numbers are, but they’re pretty well known in the industry.

Q: Is the public aware of power storage issues, or are stories of battery explosions overblown and misunderstood?

A: I think there’s the potential for HUGE confusion in the public about what “lithium-ion” means. It’s hard enough to get consumers to focus on different types of batteries: Lead-acid, nickel-cadmium, nickel-metal-hydride, lithium-ion … most people glaze over after the first hyphen. So automakers have some explaining to do, to make the case that their lithium-ion batteries aren’t the same kind as in those videos of flaming laptops you saw on YouTube.

That said, there are something like 50,000 gasoline fires a year in this country. We would NEVER approve gasoline as an onboard energy storage medium today. It’s way too dangerous! And yet you rarely see car fires on the evening news unless a family gets barbecued.

Still, GM and Toyota know perfectly well that the first minivan with lithium-ion batteries that goes up in smoke could generate a Congressional investigation. So they have to go for absolutely the safest option. Not the second-safest, or the “almost as good” chemistry, but the very best one they can identify.

Toyota may have stumbled by betting on the wrong chemistry initially. Its long-time partner Panasonic has its expertise in the cobalt dioxoide chemistries, which is (quite reasonably) what they use in consumer electronics. But those are also the ones most susceptible to runaway oxidation (fires) because they have so much oxygen bound into their chemistry.

Q: To what do you ascribe the huge increase in battery R&D? Is it because of green consciousness? Mobility? Or something else?

A: Energy storage is a desirable thing for any number of reasons. Utilities by and large don’t have it; what they make is pretty much consumed in an instant. They’d love to have it. On the other end of the scale, portable electronics—the BILLIONS of cell phones out there—demand batteries too. Heart pacemakers, solar-powered emergency callboxes, the list is endless. You can never have too much battery storage.

Q: Most of the lithium in the world is in China and Argentina. Should we be thinking about “peak lithium”?

A: Nah. Not for a long, long time. You can recycle lithium-ion car batteries, too, after their lives are exhausted in 20 years or so—just like we do lead-acid car batteries today.

Q: What’s the biggest challenge right now with batteries — charge time, battery life, number of recharges, speed of discharge?

A: All of them.

Q: Do you think the biggest steps in battery improvement will come from chemistry, such as variants on Lithium-ion or iron nanophosphate, or from physics, such as silicon nanotubes?

A: I’m nowhere near smart enough to handicap that. If I were, I’d rather be working on Sand Hill Road than on my laptop in windowless media rooms around the world! Again, though, I think it’s a combination of both. Iron-phosphate is a newer chemistry, but then there are A123’s nanophosphates—same chemistry, but major improvements in performance due to their nanostructure.

Q: Let’s go back to cars for a bit. How much are people thinking about disposal of batteries and cradle-to-grave environmental impacts of batteries in the car world?

A: The Europeans are way ahead on this one. If I remember correctly, the Germans now require automakers to take back their cars sold in the country and dismantle them for recycling.

As for batteries, the industry has been enormously successful in recycling standard 12-Volt lead-acid batteries. That’s partly because lead’s such a terrible metal to have floating around, but I think 12V car batteries may be THE most recycled consumer good—I think the figure’s 97% or 98%. The question is whether the new battery makers—they’re not the folks who make lead-acid, by and large—will hop onto that bandwagon too.

Q: How much of a gain can we get from better microprocessors and improved control over things like fuel mix, discharge rate, and computer-assisted driving? Are there parallels in the microprocessor world?

A: Enormous, huge, hard to imagine. Ford just announced its EcoBoost engine, which is direct-injection gasoline with a turbo. That V6 will probably dispense with most of the V8s in their cars, except for the vanity models (like the Mustang) where you gotta have a V8 for street cred.

We already have cars that interpret their own performance and surrounding environment, and react far quicker than any human. A surprising number of cars now ignore or modify driver commands if they calculate those commands will lead to an unsafe situation or a loss of traction.

So, suppose you could program your nav system to take not only the quickest, or the shortest, but also the most economical route? It would know altitudes as well as roads, and pull in real-time traffic data, so it would calculate total energy use from point to point. That’s easy to do, and it’ll be here within a very few years.

Then imagine your car could actually do the rush-hour stop-and-go driving FOR you, using the most economical acceleration and braking algorithms. That’s probably around 2016 or so—and it ain’t just me. That’s from GM’s head of R&D.

Q: Can a massive car like an Escalade hybrid be considered a “green” vehicle when it consumes more gas than a smaller, traditional combustion car?

A: Well, everyone has to decide what “green” means to them. To me, it’s in danger of becoming a meaningless marketing word, like “new” or “lite” or “fresh”. Ever heard of someone selling “stale”? The Escalade Hybrid is green-ER than the one sitting next to it on the floor without the batteries.

The GM party line has a certain truth to it. If you make a percentage change in the least economical vehicles—from 16 to 24 mpg, say—you’re saving more actual gasoline per mile than if you do it in a small car, going from 30 to 45 mpg. Plus, Americans buy far, far more full-size pickups and sport-utilities than they do small cars like the Prius.

To me, the question is how many of those Two-Mode Hybrids GM can afford to sell (at a loss estimated as up to $10,000 each at the start). Toyota sold 150,000 Priuses last year, plus a bunch of other hybrids. I’ve heard that GM might sell 5,000 to 10,000 Two-Modes next year. But we’ll see.

Q: Plug-in electrical vehicles require changes to the grid, whereas hybrids can generally use the existing gasoline delivery system. How big a factor is that in the success of plug-ins?

A: What changes to the grid? The manufacturers aren’t going to repeat the EV1 mistake, where you have to build a special charging system. Your Vue Two-Mode Plug-in or your Volt will plug right into your standard 110V wall plug with a 15-Amp fuse. It may not charge all that fast, but it’ll work.

Smart metering will allow time-of-day pricing, utility load balancing, and some other good stuff. But with $4+ gasoline, home charging may not seem so scary—and you know the automakers are going to say, “It takes 20 cents of gasoline to drive you a mile—or 3 cents of electricity. You do the math!”

One note, though: There are fewer than 100 plug-in hybrids today IN THE WORLD. Five years hence, there’ll be probably 10,000 or more. But I worry that people assume plug-ins are already with us. Let’s be clear: These are very complex, very challenging cars to build, because they are hugely more demanding on their batteries than today’s hybrids.

Q: Has battery technology evolved so far so fast that other approaches — like fuel cells — stopped innovating?

A: Nope. But electric-drive cars are a lot easier to integrate into life. Most Americans have garages with electric power in them. Hardly a single American can buy hydrogen fuel today for automobiles, even if the cars existed. Look at the Honda and Chevrolet test programs; they’re only rolling them out in cities where they’ve managed to install a publicly-accessible hydrogen station. The infrastructure challenges are substantial.

Plus, the power companies LOVE the idea of night-time (“off-peak”) charging for plug-in hybrids. It uses spare generating capacity, and the infrastructure’s already there. They’re all for it. There’s no hydrogen industry with one-hundredth the influence of the electric utilities.

Q: Do you wish gas cost $5 a gallon?

A: I drive in New York state, where it’s already close to $4/gallon. But I live in Manhattan, which has the lowest energy consumption per capita of anywhere in the country.

I do think it’s significant that Alan Mulally, the CEO of Ford Motor Company, said that CAFÉ requirements were the wrong way to achieve the goal of saving energy. He said that higher gasoline taxes would allow the market to decide what kind of cars it could afford. I think the answers to that are pretty clear in places like Europe. But this country seems to view taxes as roughly equivalent to bestiality, so he did say he kinda recognized it was a non-starter.



Interesting name being EV NUT, considering that you seem to be heavily biased against EV’s.

Your math may reflect the most inefficient home-built vehicle, based on low-cost components such as a DC motor and lead-acid based batteries. However an OEM would produce nothing like this and use lithium based battery technology and a more efficient AC motor.

An example of this would be the Tesla. IIRC the Tesla achieves roughly 225 watts per mile. Your math assumes that EVERYONE drives 65 miles per day. IF this were true, this would only be 14,625 watts consumed and not 25,000 or more like you suggest. Take this with a 110v (commonly outputs 120v) 15amp circuit which translates to a supply of 1800 watts per hour. This means that a simple outlet could recharge the pack in 8.2 hours. Of course as you mentioned, the charging efficiency is not 100%, nor is converting that to mechanical energy 100% either, but not to the tune of taking 10 – 12 hours to recharge. EVEN IF that were the case, I get home at 7pm and if I plugged in to charge until I leave for work at 8am the next day, I’ve got 13 hours of charging completed which falls realistically within your worst-case scenario. I am an example of a cummuter who spends over an hour commuting each day, however it consists of no more than 25 miles total with 48 traffic lights to contend with. BWT, the regenerative braking is far more than a 10% return. It is more around 60%, of course depending on how much you make use of that with your driving style. WHEN EV’s become common, there will be charging stations you would be able to connect to while at work to keep your energy level from becoming depleted.

If you would like even better evidence, I personally know an individual with a Scion eBOX (xB converted to electric by AC Propulsion) who is capable of a 120 range within city driving and performs a recharge from a standard outlet within 13 hours. Anyone like himself with an electric vehicle would be interested in installing a 240v 50amp circuit in their garage which allows him to perform the same recharge within 3 hours.

If your worried about pulling too much from the grid, the AC Propulsion controller allows him to dial-in the amount of load it pulls from the circuit it is connected to.

If you want to go beyond that, the AC Propulsion controller has existing technology that allows a smart metering system from the power company to control remotely the amount of current his charging system is pulling on the fly, in real-time. The system would be capable of communicating to the charger to variably reduce the amount of charging current, halt the charging, and if needed reverse the current and draw from the battery pack. With enough vehicles like this on a grid, if the load within the grid became too high they could simply request the from the thousands of BMS (Battery Mgmt System) units that are near full capacity to momentarily discharge back into the grid to function as a capacitor/buffer device on the electrical grid. Since energy cannot be stored in the form of alternating current and only in the form of D/C, these controllers allow the D/C energy within the vehicle to be converted on the fly into A/C energy since no power company has this ability to handle sudden loads like this. This would be a major benefit to any power company.

This technology is not even that high-tech. With the internet, let alone basic radio channels that could be tuned into, this could be accomplished easily. Many communities have radio channels that are tied into local weather stations and can tell your water sprinkler system when to water the lawn. In south Florida (at least) FPL has had ON-CALL BOXES installed into nearly all residential for about 15 years now. These boxes are wired into high-load appliances such as water heaters, sprinkler pumps, etc. and FPL can remotely choose to turn these systems off unit-by-unit if the load on the grid is becoming to high, permitting relief to the grid when needed.


john harry nickelson

which is the battery suitable for 1kw range motor running in an electric vehicle,the battery should last for 4years….?

Ray C.

I get a big kick out of the nah sayers when they talk about how lacking the infastructures are if we were to change over to all electric vehicles. Its not like the change would be overnight with 200,00 vehicles suddenly plugged into the grid. This would be a gradual shift allowing time for the electric producers to upgrade their production and delevery systems. Somehow the automakers must be convienced that electric is the way to go and let free enterprise run its course. If the oil companies want to raise the price of gas to $4+/gallon, then we need an alternative way to use less of it. If it is only commuting short trips or back and forth to work. Just think what would happen if the consumption of gas went down just 10% in one year, what kind of message would be sent.


Green energy is definitely the best solution in most cases. Technology like solar energy, wind power, fuel cells, zaps electric vehicles, EV hybrids, etc have come so far recently. Green energy even costs way less than oil and gas in many cases.

Daniel Kongs


You seem to think technology and knowledge are static and the rules for today are the way it will always be. I charge my electric car very nicely with solar panels – technology still developing and not even thought of just a few years ago. What is your answer-find more oil?



OK, so maybe you go for a 3-Phase plug 240V, 30 amp. That would reduce the charge time to about 5 hours. If you had a charge circuit at your workplace, you could reduce the overnight charge time further. These are not insurmountable problems.


Plugging in an electric only car overnight only works if you’ve driven for less than 1/2 an hour, you’re driving a go-kart, or you have a new power circuit installed.

I haven’t done a survey of how happy electricity utility companies are with the idea of lots of people charging their cars overnight, but I’m a little worried that the math may be grossly underestimated.

It takes about 20kW of power to run a small car on a highway at 65MPH, for an hour. Stop and go traffic (city driving) is typically about 25% (to 35%) worse than that (Don’t bother betting on regenerative breaking to save you – you get at most 10% of your energy back into the batteries).
According to the US census, the national average American commute is about 25 minutes each way. Let’s just say 60 minutes per day total, for convenience sake.

That’s 25 kW of power taken every day per electric-only car that needs to be charged each night.

At 80% charge efficiency (you need more electrons to charge a battery than you can extract from it later) and 90% AC-DC power conversion, you’ll need 33kW of power each night to charge your car.

problem 1: Given a 110v, 15A circuit, it will take 20 hours to recharge your car if you can run your 110v, 15A circuit right at the max. Even if you sell the idea as “It’s cheap!” as in your example, it just doesn’t work – you’d need to charge in 10-12 hours or less, or your car won’t be ready in the morning.

Problem 2: Especially in colder climates where electric heating is still in use, don’t expect that an extra 33kW for a 10 hour period in every household for every weeknight will not require major upgrades to the power grid. Taking a base example, the average US house draws just over 10,000 kW hours in a YEAR. Given this new proposed electric car demand, just for weekdays (assuming people stay home on weekends, or take public transit) this is a 66% power demand increase, on average. If you’re a power company, and need to plan for peak and not average, the numbers look considerably more scary. So, I don’t for a second believe that power grid infrastructure will just absorb this without some major upgrades, even in this “best-case” scenario where I assume your average American only uses a car 5 hours per week.

ev nut

the well known numbers are lithium is way to expensive. vaporware a or for the rich. nickel metal hydride would easily run the chevy volt 40 miles then the ice generator gets you as far as you want to go. but nimh has been suppressed by gm/chevron/cobasys just when we need it most. shame


Q: How efficient do you think batteries will become in five years’ time? Why?

A: The improvement in batteries’ energy density runs on a fairly steady curve, and so does the decline in cost per kWh. Think of it as a battery version of Moore’s Law. I couldn’t tell you off the top of my head what the numbers are, but they’re pretty well known in the industry.

Can you follow up on this? I would be interested to know what the answer is. It may be well known in the industry, but it isn’t common knowledge outside it (unlike Moore’s Law).


Comments are closed.