January 08, 2010
Boston Consulting Group On Electric Car Battery Costs

Boston Consulting Group says car battery costs will not fall far enough in the next 10 years to allow a massive shift toward electric vehicles.

Although electric-car battery costs are expected to fall sharply over the coming decade, they are unlikely to drop enough to spark widespread adoption of fully electric vehicles without a major breakthrough in battery technology, according to a new study by The Boston Consulting Group (BCG).

The study, released today, concludes that the long-term cost target used by many carmakers in planning their future fleets of electric cars--$250 per kilowatt-hour (kWh)--is unlikely to be achieved unless there is a major breakthrough in battery chemistry that substantially increases the energy a battery can store without significantly increasing the cost of either battery materials or the manufacturing process.

"Given current technology options, we see substantial challenges to achieving this goal by 2020," said Xavier Mosquet, Detroit-based leader of BCG's global automotive practice and a coauthor of the study. "For years, people have been saying that one of the keys to reducing our dependency on fossil fuels is the electrification of the vehicle fleet. The reality is, electric-car batteries are both too expensive and too technologically limited for this to happen in the foreseeable future."

BCG isn't just saying cost is a problem. They also see weight as holding back EVs. That makes sense. The 400+ lb battery in the Chevy Volt provides a 40 mile range on electric power. To go 200 miles in electric power would require 2000 lb just for the battery. Forget about the typical car's 400+ mile range until battery energy density goes up by some multiple.

BCG says currently prices are between $1,000 and $1,200 per kwh. To put that in perspective a compact or midsize car might use a quarter of a kwh per mile. So at current prices a 100 mile range will require 25 kwh or at least $25,000. The cost is worse than that since batteries are not typically allowed to run all the way down.

Most electric cars in the new decade will use lithium-ion batteries, which are lighter and more powerful than the nickel-metal hydride (NiMH) batteries used today in hybrids like the Toyota Prius. Citing the current cost of similar lithium-ion batteries used in consumer electronics (about $250 to $400 per kWh), many original-equipment manufacturers (OEMs) hope that the cost of an automotive lithium-ion battery pack will fall from its current price of between $1,000 and $1,200 per kWh to between $250 and $500 per kWh at scaled production. BCG, however, points out that consumer batteries are simpler than car batteries and must meet significantly less demanding requirements, especially regarding safety and life span. So actual battery costs will likely be higher than what carmakers predict.

The OEMs are hoping for a substantial price decline once volumes go up. Will that happen in just a few years?

BCG talked to many major players to come up with their cost numbers. Any optimists want to dismiss this report in the comments section? (you know who you are)

The report, titled Batteries for Electric Cars: Challenges, Opportunities and the Outlook to 2020, is a companion piece to a report BCG published in January 2009 on the future of alternative power-train technologies (The Comeback of the Electric Car? How Real, How Soon, and What Must Happen Next). The new report's findings are based on a detailed analysis of existing e-car battery research and interviews with more than 50 battery suppliers, auto OEMs, university researchers, start-up battery-technology companies, and government agencies across Asia, the United States, and Western Europe. The report also draws on the firm's extensive work with auto OEMs and suppliers worldwide.

BCG's numbers seem hard to dismiss.

Only $360-$440 per kWh by 2020 if BCG gets it right.

To show how battery costs will decline, BCG uses the example of a typical supplier of lithium-nickel-cobalt-aluminum (NCA) batteries--one of the most prominent technologies for automotive applications. BCG's analysis suggests that by 2020, the price that OEMs pay for NCA batteries will decrease by 60 to 65 percent, from current levels of $990-$1,220 per kWh to $360-$440 per kWh. So the cost for a 15-kWh NCA range-extender pack would fall from around $16,000 to about $6,000. The price to consumers will similarly fall, from $1,400-$1,800 per kWh to $570-$700 per kWh--or $8,000-$10,000 for the same pack.

Batteries could still conceivably go into wider use in 2020 if the availability of oil becomes so limited that pluggable hybrids become the preferred way to get to work. 40 mile range on electric power ala the Chevy Volt would allow most people to do their commutes without gasoline. Pure electric cars with 100+ mile range are going to maintain more of a niche status unless prices fall even farther.

Note that a premium electric sports car maker can sell an electric car with batteries that do not last as long as mass market customers expect. So an exotic sports car maker can use lower priced lithium batteries designed for computers. But a company like General Motors needs to achieve a much higher durability and reliability in a mass market design.

Update: The real cost of electric car batteries continues to be debated on web logs and in the press. Back in March 2009 Jon Lauckner of GM criticized a CMU study on electric cars by claiming that $1000 per kwh is hundreds of dollars too high. So why is BCG, which certainly knows about GM's claim, citing a higher figure?

At its core, the study’s conclusion is based on an incorrect assumption of the cost of battery packs. In the CMU study, the so-called “base case” used a Lithium-Ion battery cost of $1,000 per kWh ($16,000 for a 40 mile Volt pack) that was cited in earlier academic articles. The problem is this cost is many hundreds of dollars per kWh higher than the actual cost of the Volt pack today. Moreover, our battery team is already starting work on new concepts that will further decrease the cost of the Volt battery pack quite substantially in a second-generation Volt pack. Unfortunately, the impact of dramatically lower battery costs (to $250 per kWh) was treated only as a “sensitivity” in the CMU study when it probably should have been highlighted as THE critical element that would dramatically change the cost-effectiveness of plug-ins with greater electric-only range.

What's behind these conflicting prices on EV batteries?

Update II: Here is the new study (PDF).

Share |      Randall Parker, 2010 January 08 08:08 PM  Energy Transportation

Fat Man said at January 8, 2010 9:12 PM:

I am not surprised, although I am not as optimistic as BCG is about BEVs.

Isegoria said at January 9, 2010 6:43 AM:

As we've discussed before, batteries are extremely expensive, not just in dollars per kWh, but also in pounds per kWh, even if the electricity they store is fairly inexpensive per kWh. Gasoline stores tremendous energy for its weight.

Thus, if you're going to include both a gasoline generator and batteries, as in a Chevy Volt, you should only carry enough electricity in battery form to get you to work (and maybe back, if you can't charge there). Any storage you don't need on a more-or-less daily basis is a waste of money and a waste of energy, because those extra batteries don't carry themselves.

simone said at January 9, 2010 7:53 AM:

I think the easiest way to assess BCGs claim is to see the curve they have assembled (or would assemble) to characterize the development of battery technology. We can then see if the technology is evolving in a linear or exponential manner. My quick read of the blurb leaves me wondering what they really did to develop the forecast.

My casual reading of this space suggests that it is moving very rapidly. I would not rule out the possibility of more rapid progress.

With that said, the amount of oil and gas is sufficient on the planet to easily accommodate world demand for the next 40 years. Our only concern should be politicians and fools who wish to keep humanity in the constant state of fear to ensure their dominion over us. Let's take a deep breadth and accpet we can build a brighter future.

Reid of America said at January 9, 2010 8:18 AM:

I predict that a breakthrough in superconducting capacitors will make batteries obsolete. They would be much lighter and not degrade over time like batteries.

K said at January 9, 2010 11:34 AM:

I like the BCG presentation because it separates battery cell costs from battery pack costs. And it does not speculate on what new cell chemistry may emerge.

It projects that even if the costs of cells fell to $0 the pack will still cost $90 - $110 by 2020. Roughly 25% of the 2009 cost. So to reach the $250/kWH Grail the costs of cells must come down to roughly $150/kWH.

i.e. decline about 80%.

The projection of pack costs seems likely to be close. AFAIK packs require no rare materials or new techniques, cutting cost is mostly a matter of mass production.

The question is whether the cell costs can be reduced by 80% by 2020. Or will they fall by about 60% as this report suggests.

That GM criticism from March 2009 may not conflict with the BCG. It does not say what volume GM uses when projecting pack cost or give the years of production for either first or second generation packs. Pack production has just started. Volt production is announced for November 2010.

The fog is lifting. I wish the prospects of fuel cell vehicles were as clear.

Nick G said at January 9, 2010 12:39 PM:

Unfortunately, we don't have the study - I've requested a copy, but at the moment we just have the press release. So, we can't do a serious review yet.

Preliminary review:

1) the study agrees that 1st generation li-ion costs $325 +/-$75, slightly less than the $350 figure I've been using for both 1st & 2nd gen cells,

2) it projects fairly strong growth of 25%/year (3% market share growing to 26% in 2020),

3) 2nd gen li-ion should cost the same or less to manufacture, so the roughly 3x factor very likely refers to the combination of "inactive" reserve to reduce depth of discharge and the overhead cost the Balance of System (battery management system, temperature control, complex packaging ). This fits perfectly with the explanation of battery costs from the CEO from CPI (the company that builds the Volt packs today), who says the pack will cost $350/KWH for the cells, and $1,000 per KWH for the available 8KWH capacity of the battery pack, which equals $8,000 per pack.

"All four of these items together justify a 2.5x premium for the AT application (or approximately $ 1,000/available kWh) compared to the $350/stated kWh of a CE system, CPI says."


So, in fact we seem to have confirmation of the previous battery cost info from GM and Nissan.

Engineer-Poet said at January 9, 2010 7:54 PM:

Battery-pack cost per kWh isn't the only variable; Better Place throws in another, which is the number of kWh per car.

What fraction of cars are on long-distance trips at any time?  If only those cars are carrying high-capacity packs (10-30 kWh and up) and most people are using 5-10 kWh packs for tooling around town, the number of high-capacity packs and their cost will be much smaller.  Low-capacity packs in the same form factor have radically reduced constraints for energy density, so should be much cheaper and can use completely different chemistries.  This reduces the overall capital investment and still achieves complete electrification.

Nick G said at January 10, 2010 8:35 AM:


Good point - anyone who wants to evaluate Nissan's business plan better include Better Place, and the demand it's going to create in Denmark, Israel, San Francisco, etc. Those governments recognize the social costs of an oil addiction, and are making a large commitment to building charging stations to solve the range problem, and buying EV's from Nissan.

Bruce said at January 10, 2010 9:03 AM:

During a webchat this week, GM vice chairman Bob Lutz said again that higher gas prices are needed for the Volt to become "generalized." Specifically, he said:

The Volt technology is very exciting, but costs will have to come down before it can become generalized, and U.S. fuel prices will have to rise to world levels, meaning $5 or $6 per gallon.


OUCH! Batteries are going to be 1000$/kWh

Pink Pig said at January 10, 2010 11:55 AM:

That was quick. Another pipe dream bites the dust. I'm waiting for someone else to notice that batteries need recharging with electricity, and that electricity is mainly produced by burning coal. It may well turn out that battery-operated cars need more fossil fuel energy (!) than gasoline.

Bruce said at January 10, 2010 12:21 PM:

"In 2009, lithium carbonate prices jumped to $6,500 a metric ton, almost tripling 2006 values, because of surging demand for batteries, Jaskula says."


Chris said at January 10, 2010 12:56 PM:

Two things need to happen for EVs to actually take off:

1. They need to be designed and built to do more than drive your typical latte-sipping urbanite down to pick up his 2-3 best buds for their latest trip to the nightclub.

2. They need to be demonstrated to normal folks (i.e., the majority of drivers who do not live in an urban setting) that they are not a tremendous step BACK in power, performance, and range, for a tremendous step UP in cost. In short, too many "green" solutions suck. R-134a replacing R-12...the total elimination of DDT use worldwide (even where minimal use would provide termendous gains)...there are so many other examples. When will we learn that we can't just kill good technologies without coming up with a COMPARABLE replacement???

So far, that's all that EVs are good for. Far too many are spending far too much time telling us all what we "should" be driving, without telling us why (aside from "you're saving the planet"). No one is coming up with a transitional strategy from carbon to whatever, and this report demonstrates further that while it might be possible to achieve great gains down the road, we're nowhere close with the new stuff just yet.

@Reid of America: there's been a lot of serious research into superconductors for decades (I myself remember hearing often of the promise of superconductors when I was in junior high in the 80s). Yet you might as well wish for a revolutionary battery technology or hydrogen fuel-cells...the sort of serious superconducting capacitors you're talking about are just as far off as anything else we're wishing for here, so why not say "Portable fusion power is right around the corner...and that will eliminate all of these concerns?"

Caveat: I'm all for fusion power development, fuel cells and other portable power technologies, nanotech-based batteries, superconductor use...but you can't force innovation. What that means is, rather than pushing out half-baked solutions that suck really badly and just give the new stuff a bad name, why not find innovation in what we have, continue to make gains there, while FULLY developing the new replacement technology? Trust the business models and the natural progression of science; there's a good reason why we're not all taking the train cross-country today or sailing the oceans on steamships to Europe, you know. There IS money in all this, and the stuff WILL be developed...let's just not delude ourselves that we're doing it in 5 years (possible but not at all likely). We've all gotten too used to scientific development on Internet time, where innovation can happen in months and new cool products come out almost every week.

Nick G said at January 10, 2010 1:29 PM:


1) The second generation of the Volt will be rather less expensive, as economies of scale, and longer time-lines for design improvement and competitive procurement kick in. That's part of what Lutz is talking about.

2) Despite the fact that Lutz came up with the idea for the Volt, he's really an old ICE gear-head at heart, designing things like the Viper. He flies a personal jet - he likes speed, and he associates that with ICE's.

3) A lot of people feel the same way. There's an old rule to innovation: something new can't be as good as the old thing, or even 2x as good - it has to be 10x as good to overcome the inertia of the status quo. Even though electric motors fundamentally give better performance with low-speed instant torque, a lot of people will need a lot of convincing.

So, yes, it will take a while, and improved pricing, for the Volt to become "generalized."

Pink Pig,

Yes, EV's are primarily a solution for Peak Oil not for CO2 emssions. On the one hand, that good enough. OTOH, EV's provide night time demand (which both wind and nuclear power need) and schedulable charging to soak up wind's intermittency, so EV's really do help with CO2 emssions.


EV's have better performance than ICE's. The Tesla has better performance than ICE sports cars that are several times it's price, and the Volt is 0-60 in 8.5 seconds.

Mike McDaniel said at January 10, 2010 1:48 PM:

OK, let's think a bit now. BCG claims $16,000.00 for a 40 mile range battery for the Chevy Volt, but GM says "The problem is this cost is many hundreds of dollars per kWh higher than the actual cost of the Volt pack today." Oh yes, and they also say " Moreover, our battery team is already starting work on new concepts that will further decrease the cost of the Volt battery pack quite substantially in a second-generation Volt pack."

So, GM claims their batteries will be "many hundreds of dollars less." Uh-huh. So that's not a thousand less, but "many hundreds" less. So instead of $16,000.00, a battery pack will cost what? $15,500? $15,300? That's still more--for just a battery(!)--than the price of some conventionally powered vehicles currently on the market. While I commend GM for not exaggerating, this is hardly an improvement that is going to make electric vehicles a viable rather than a tiny niche product. And I'm glad GM is working on the battery issue, but they're "starting work on new concepts," not completing work on a scientific breakthrough that will be orders of magnitude better than current technology. Until then, BCG is squarely on the side of reality and GM is on the side of Obama/green fantasy.

And while we're talking about batteries, I'll be very interested to hear how GM is planning on repealing the laws of physics such that battery powered vehicles will be viable in any part of the US where it actually gets cold, cold having the tragic effect of rapidly depleting the charge of any known battery. Perhaps they're starting work on new concepts in that regard too. I see almost no one talking about this, but the simple fact of cold and its effect on battery charge and capacity is what always has, and always will, absent dramatic scientific breakthroughs, put a stake through the heart of electric vehicles.

All the wondrous "green" solutions in the world are meaningless if they're not at least as efficient, if they don't provide at least as much value and flexibility, as current technologies. Let's not even mention how useless green goodies are if no one can afford them.

Nick G said at January 10, 2010 1:57 PM:


BCG didn't claim $16,000 for a 40 mile range battery for the Chevy Volt. They claimed $1,000/KWH for the effective 8KWH capacity, or $8,000.

GM claims their batteries will be "many hundreds of dollars less."

That's per KWH, not per battery pack.

The problem of cold is not that hard to solve. Remember, conventional batteries have this problem, and so do conventional ICE vehicles. The answers: garages, and block heaters (or the EV equivalent). An EV with a very large battery will have some reserve even in very cold temps, and will warm use quickly with use.

OTOH, a well designed extended range EV (aka plug-in hybrid), like the Volt, has a climate control/temperature management system for the battery.

A plug-in hybrid is designed to be....plugged in. That means that most of the time the minor power needs of the temp management system will come from the grid. For the small fraction of the time that one is parked out in the cold for an extended period, the ICE generator will start immediately when the car is started, and provide power during the short time that is required to bring the battery up to optimal operating temperature.

Another reason why ErEV/PHEVS seems likely to be the optimal design for most people for quite some time.

Randall Parker said at January 10, 2010 3:06 PM:


EVs could take off if more people sipped lattes. What's needed is a big lifestyle change.


Even the battery pack size of a Chevy Volt with a battery sized for a 1 day commute round-trip is too expensive.

Nick G,

Denmark and Israel are small places. Their car usage is probably much less than a single year's demand growth in China. Scaling up the rental/leasing battery model to the US market is a financially far tougher job that isn't going to work without gasoline going for something well above $5 per gallon. Conventional hybrids are a better deal. EVs will have a hard time competing with HEVs at anything less than $10 per gallon.

I look at it this way: The Prius is the target to beat. It can go 12,000 miles on 240 gallons of gasoline. At what price of gasoline does 240 gallons per year cost more than a PHEV battery's cost plus the cost of electricity? At $5 per gallon 240 gallons is $1200 per year or $100 per month for gasoline. That's a tough number to beat for a PHEV's incremmentally higher cost of battery, electricity, and other costs.

A key question: What percentage of a typical car's usage in a year would be done on electricity given a 40 mile battery range?

Micha Elyi said at January 10, 2010 5:55 PM:

Bruce (9:03 AM) wrote:
"...U.S. fuel prices will have to rise to world levels, meaning $5 or $6 per gallon."

U.S. fuel prices are already at "world levels." It's U.S. fuel taxes that aren't at "world levels."

Just because something in the U.S. isn't at "world levels" doesn't mean America is doing something wrong. Would you like "world levels" of dictatorial government? Of poverty? Of disease? (The adults here sure don't.)

Nick G said at January 10, 2010 6:29 PM:

What percentage of a typical car's usage in a year would be done on electricity given a 40 mile battery range?

78%. I'll look for the reference - IIRC, it was DOE driving pattern data. It assumes only 1 charge per day, and no modifications of driving patterns to maximize electric driving.

That gives 9,360 electric miles, and 2,640 gas miles. At $.10/KWH (which is conservatively high), 40 MPG (conservatively low), $3/gallon and .25KWH/mile, that gives an operating cost of $432, and savings of $468/year.

Of course, it's important to remember that the 12K miles is an average. Some people, like me, who only drive 2k miles per year (I live in a big city, my house has a walkability index of 92, and I commute via electric train) will have a very hard justifying almost any capex to reduce gas consumption. OTOH, at least 10% of VMT goes to drivers who go over 20K miles per year, and many drivers will have access to supplemental charging during the day.

Denmark and Israel are small places.

Sure, but there's a pretty good chance that they, and other locations like San Francisco, will produce a locked-in market. That's very different from a traditional marketplace of the kind analyzed by BCG, and leaving those volumes out skews the analysis. If Nissan has a substantial level of guaranteed sales, it makes a real difference to their business model.

The Prius is the target to beat.

The Prius is a great car. Further, it can be upgraded to a plug-in for as little as $3,500 (using lead-acid) http://www.3prongpower.com/ . The Prius plug-in due in a couple of years will have a very sweet spot, cost-wise.

On the other hand:

1) the Volt will have much better performance (0-60 in 8.5, vs 10 seconds IIRC), and it's greater weight and lower center of gravity give a better driving experience. So, the two aren't comparable.

2) The Volt will get a $7,500 tax credit for the first 250,000 vehicles. This is at least partially justified by the real external costs of imported oil. That is likely to make it's life-cycle cost comparable to a Prius.

3)After the credit expires there will be a % of buyers who are willing to recognize those external costs, even if they're not out of pocket - that % is small, but real. Heck - if everyone was just trying to minimize their out-of-pocket cost per mile, we'd all be driving $12K, 35MPG Korean sub-compacts, right?

4) Battery costs should be amortized using a capital ratio of 10:1. The 10-year Volt battery life is a minimum - it will be very likely, on average, to perform to specs for a substantial period past the 10 year warranty period, and decline gracefully after that point.

In 4-10 years we can expect the Volt to have only a $5K premium over a Prius, which would make it's life-cycle costs comparable at $3/gallon for the average vehicle, much lower for above-average mileage drivers. Of course, there's a pretty good chance gas will cost more than that...

Chris said at January 10, 2010 7:05 PM:

To Nick G:

It carries two people and zero cargo (unless your maximum cargo needs fit in a briefcase). There are also cars that cost much LESS, come close to matching those numbers, carry more, and drive farther without stopping. The Tesla, while cool-looking and impressive, is also specially-designed. This is fine, but why can't Tesla spend their money and development time designing a car for normal people, that does what normal people need?

To Randall Parker:

Another point made for me: "the car you SHOULD be driving..." along with "the lifestyle you SHOULD have..." Well done. You've TOTALLY convinced me!! Although it's pretty hard to hear you with you WAY up there on your high horse...I need you to speak up a little so I can hear you patronizing me.

Seriously, though, what's REALLY needed is not a lifestyle change...it's progressive technology that is indeed sustainable, yet permits freedom of choice to all individuals to continue their inalienable right of "the pursuit of happiness" and doesn't REQUIRE a lifestyle change. (Are you all saying that it can't be done?) All you folks only look at the extreme ends of this situation...the Tesla is really fast, but it's totally impractical. As to the Tesla, I wouldn't buy a Ferrari, either, except as a very expensive toy...just like the Tesla. If rich people want to buy it, they will. The Volt is an interesting concept and certainly (theoretically) practical, but as this report shows, it isn't practical now and won't be for some time. Mainstream consumers won't want to buy these until they come reasonably close, say, within about 85-90%, to the FULL capability of a regular old-school vehicle and cost perhaps 5-10% more at most (and preferably the same or less). I'm not here to just argue against EVs...they're very promising as a technology and have great benefits in many ways...WHEN THEY'RE FULLY-BAKED. Don't pretend that a) they're ready for prime-time and full across-the-world adoption TODAY, or b) the critical breakthrough to do so is right around the corner.

Let me provide one last analogy: the first serious automobiles were built in the 1880s. The first AFFORDABLE (and thus capable of becoming common) automobile here in the US was the Model T in the late 1910s and 20s...so roughly 30 years. And the Model T was still pretty primitive. The EV is a major paradigm shift in the automotive field...do you somehow expect it to succeed in a mere half-decade or so? We have advanced in ways that might encourage quicker breakthroughs from how they came 100 years ago, but breakthroughs in physics do not yet come on Internet time (as I said earlier). We're in the 1880s of EVs now. Even if it takes HALF the time of early automotive development to reach reasonable REAL costs (not subsidized costs, as the early autos had no government to subsidize them) and mainstream commonality, you're still looking at approximately 15 years...FROM TODAY. So again, we're matching up well with what this report is claiming.

Randall Parker said at January 10, 2010 8:25 PM:


I was being sarcastic when I said people should drink lattes in order to want EVs in their current state. Try reading for levity.

I didn't tell you which car you should be driving.

No, I do not expect EVs to become competitive in a half decade. This is GM's 3rd EV development effort. The EV-1 was the second attempt according to a retired GM manager I talked to who worked on the first two designs. So it is taking decades.

Actually, before oil was discovered in large quantities in Texas in 1894 electric cars were as popular as ICE cars. Cheap gasoline shifted the terrain in favor of ICEs. So electric car development stretches back over a century.

Dan said at January 10, 2010 11:05 PM:

Jumping back to the original question in the post: Is it reasonable to assume battery prices for PHEV applications will fall from the current price of ~$1000/kWhH to around $250/kWh in the near term?

I'm a bit of an optimist. The fact that batteries for CE devices can already be purchased for $250/kWh implies that economies of scale can drive the manufacturing costs to appropriate levels if technological change is sufficient to overcome the difference in requirements between automotive and CE device applications. This is quite a bit lower bar than requiring fundamental advances in chemistry or physics. The main differences between batteries for vehicles vs cellphones is in calendar life, deep-discharge performance, safety, and prismatic form-factor.

Newer chemistries (e.g. Panasonic's new offerings) have practically eliminated safety / heating issues. There is quite a bit of successful research into improving the useful discharge range and lifetime of batteries which does not require new physics; just incremental improvements to anode/cathode chemistry .... and we're starting to see demo cells now which prove out many of these technologies/techniques (I saw some at CES this week). I don't have a background in primatic cells, but I expect that the costs are mostly driven by volume -- anyone know why they are more expensive? are they actually required (cf/Tesla)?


srp said at January 11, 2010 1:45 AM:

You could go with the late Petr Beckmann's suggestion: Put induction coils under the major roads and design cars that draw power from them directly, with batteries used only for driving on secondary streets. Then we can get the economies of mass power distribution without giving up the freedom of individual auto travel. No battery breakthroughs needed.

I'm not really a fan of this sort of thing--I think ICEs are just fine and will continue to improve--but if you want to go electric, the battery bottleneck is a huge problem that Beckmann's approach would avoid.

Engineer-Poet said at January 11, 2010 6:48 AM:

Induction coils would require complete reconstruction of every affected lane, and probably drive up maintenance costs a great deal too.  This is before you consider issues like EMF effects on humans (and degaussing your remaining audio cassettes and the mag stripes on the cards in your wallet).

It makes more sense to carry power through conduction.  Put the vehicles on rails (like railroad maintenance trucks) to eliminate the pavement entirely and provide a ground path, and supply power by overhead wire or center rail that is only "hot" when a vehicle is over it.  A 2003 paper claims rails are $267,000/mile and concrete ties are $190 each.  This gets you a system which can carry maximum-weight heavy trucks without suffering pavement damage, complete electrification and also automated guidance.  The labor savings from being able to hook a bunch of trucks into a train with one "engineer" in front doing all the control for e.g. overnight travel would be enormous; drivers on cross-country routes could sleep (and presumably get partial pay) while the truck was in motion.

The all-up weight of a Caterpillar C9 is on the order of 1000 kg, and it produces 575 HP in marine trim.  If we assume an electric motor with the same power/weight ratio as the Telsa roadster (200 HP/70 lb, or 2.8 HP/lb) the weight of the equivalent motor would be about 100 kg (also with a lighter transmission).  If the balance of the 900 kg savings was used to carry Zebra cells at 90 Wh/kg, the truck would carry 81 kWh of batteries; at the Balquon figure of 4.4 kWH/mi loaded, this would allow a one-way drive of about 15 miles from the rail head with reserves, or 7.5 miles round trip.  Diesel sustainers could increase that if necessary.

How much of the population and industry of the USA is within 15 miles of a rail line?  I suspect it is a very large majority.

CyclemotorEngineer said at January 11, 2010 7:59 AM:

At present battery costs, replacing a car with an electric bicycle at least some of the time makes perfect sense. For the time being, any battery powered vehicle will have to shed some armor to be competitive with gasoline powered vehicles. According to our federal guidelines, a vehicle without heavy armor is no longer a car, but a motorcycle.

Please see my Google Knol for info about life cycle costs of electric bicycles:

LL said at January 11, 2010 8:27 AM:

Somehow I dont think the all electic car matters all that much. I agree with what Randall said that the Prius is the target to beat. Guess what? I can guarantee that it will be beaten in 4-5 years....by the next generation Prius!

One of the main reasons that the ICE has been with us for so long is that there have been incremental improvement over time that makes it a great technology. Some of the improvements is from transmission improvements and other cool tricks like cylinder deactivation.

Right now there are some really good hybrid products out there:

Small car:
Honda Insight
Toyota Prius
Honda Civic Hybrid

Mid size car:
Ford Fusion Hybrid

Small SUV:
Ford Esacape Hybrid

There are also very good Diesel products out there. What we will find is that incremental improvements to all of the above will make the pure electic a hard sell.

In a few years the products above with face stiff competition from other car makers, and there will be new versions that capitalize on Improved ICE for gas and Diesel, diesel hybrids, better transmissions, incrementally better batteries (a revolution would be great but incremental improvements are fine), I'm sure the next generation would include plug in hybrids that would inch up the overall MPG even further.

I'm sure that this next generation will be here soon, within 2-5 years. By the Volt is available, its competition will be all the more tougher. Pure electic vehicles are viable in the very small range such as Golf carts and Electic Bikes etc. Of course we will see incremental improvements in these fronts as well.

BR said at January 11, 2010 10:07 AM:

The one thing I'd like to see if they took into account is the effect of energy density on range. Here's a graph of energy density (volumetric density, but it was the first thing I found) since 1985:


If the energy density continues to increase the pack gets lighter (therefore the car gets lighter), and a lighter vehicle needs considerably less energy to move (at lower speeds). This translates into a higher mile per kWh value. Essentially it isn't an algebra problem anymore it is a calculus one.

Nick G said at January 11, 2010 3:50 PM:


It carries two people and zero cargo

Uhmm..isn't that pretty standard for expensive high performance sports cars?

There are also cars that cost much LESS, come close to matching those numbers

But they don't beat them? The only ICE cars that can cost much, much more?

why can't Tesla spend their money and development time designing a car for normal people

That's always been their business plan - that car is coming out relatively soon, I believe.

My point? That EV's have better performance than comparable ICE vehicles. The same is true of the Volt - it has no operational disadvantages at all. In fact, it has much better performance than, say, a Prius, and it's dual fuel feature means that it's a great emergency vehicle during gas shortages.

Bruce said at January 11, 2010 4:18 PM:

One of the disadvantages of a plug-in hybrid or EV is that you need a wall socket.

I live in a neigborhood where most cars are parked on the street. Do you think extensions cords trailing over the sidewalk to an outdoor plug on the house will last long?

How about parkades at work and at home? How does the apartment owner meter your non-existant individual plug?

Nick G said at January 11, 2010 4:42 PM:


The EV-1 was the second attempt

I suggest a different way of looking at it. The EV-1 was successful on it's own term: it met it's design specs, and it never had a chance to achieve reasonable costs because it was abandoned in the prototype phase: GM simply didn't want to build it. Why? Because they (including investors and staff) didn't want to make their existing products (and investments and staff) obsolete. It's the same reason IBM didn't build a PC until forced to by Apple - they didn't want internal competition. GM was pushed to make an EV by Tesla and Toyota in exactly the same way.

This is GM's 3rd EV development effort.

Far more than that. GM sold EVs in the 1910's: it sold thousands of electric trucks from 1912 to 1918. These were commercial, on the road vehicles, not a pilot or a prototype. GM has designed many other EVs and hybrids over the years. The fact is that electric drive trains are very, very simple. Heck, I built electric motors and EVs from wire and wood when I was 11 (try that with a gas engine!).

As you note, EVs were commercially successful before ICE's. They had range & charging-time problems, and had to compete with dirt-cheap oil. Well, the ErEV concept solves the range & charging-time problems, and we no longer have dirt-cheap oil. As a bonus, we have new battery chemistries that make EVs more convenient to design and use, and lower cost.

What GM is doing with the Volt is spending a billion dollars to optimize an ErEV that uses the latest engineering, to make it successful as possible. It just needs economies of scale to make it cost competitive.

Bruce said at January 11, 2010 5:15 PM:

GM can't build an ICE car that makes money. Why think they can build a an EV that makes money?

Nick G said at January 11, 2010 5:20 PM:

I think it's important to get a good perspective on electric motors vs infernal combustion engines.

There are far more electric motors in the world than ICE's. They're in your washer & dryer, your food processor, your electric toothbrush, your fridge and A/C, the sump pump in your basement, etc, etc.

There are millions of EVs in the world: forklifts, golf carts, etc, etc.

The electric drive train in diesel/battery submarines, around for 90 years (in a very high-reliability and safety demanding environment), is identical to that in a Volt. The drive train of a diesel freight train (sold in large numbers by GM's Allison Chalmers division) is very similar. The largest container-ship in the world uses 80MW electric motors.

The design and construction of electric drive-trains like those used in the Volt are very, very well understood. There is nothing experimental or unproven about them.

Randall Parker said at January 11, 2010 7:38 PM:

Bruce, I've made the same point to Nick G about the problems with home recharging. In my neighborhood very few people park their car in a garage. Even those who have garages do not use them for cars. Mostly the garages are full of stuff. Many of the garages are only 1 car wide and families with 2-3 cars park on the street or in driveways.

In areas where land is expensive garages are less common. In areas where people are really poor, again not so many garages.

I'd like to see some good survey data on the percentage of cars parked in garages. Maybe I should put a survey on my site where people can vote on whether they park in a garage.

Seems to me that garage parking is necessary for recharging during inclement weather.

Nick G said at January 11, 2010 10:49 PM:

More than 90% of vehicles have garage parking or a driveway, parking lot or other off-street space.

"PARKING: Slightly more than nine in ten American households (91 percent) have at least one car, van, or light truck at home for personal use.
Because 71 percent of homeowners and 35 percent of renters have more than one vehicle, parking space can be a real concern. Garages or carports are common for households living in single-detached units—just over three in four of these homes (76 percent) have a covered shelter for vehicles. Townhouses or row houses, on the other hand, include a garage or carport less than half the time (46 percent). In both mobile homes and units in multiunit buildings, the proportion is 26 percent.
At homes without a garage or carport available, vehicles may be left either on the street or in a driveway, parking lot or other off-street space. For homes without a garage or carport, some kind of off-street space is available at 87 percent of the detached units, at about 75 percent of both the single-attached units and units in multiunit structures, and at 90 percent of the mobile homes.
All this leaves about 7.8 million households who must rely on street parking. Of course, not all of those households have vehicles. Four in ten households who report no offstreet or garage parking also have no vehicles."

Here's the source.

I live in SoCal - most people I know have too much stuff in their garages to use them. On the street where I live almost all the cars are parked on the street or in the driveway in front of the garage. What about us?

I see this occasionally, but in the midwest this is fairly rare. I would guess that it's a symptom of very temperate weather combined with home prices much higher than average for the country that put space at a premium. I would guess that the availability of PHEV/EV's will encourage more people to actually use their garages.

Apartment dwellers face similar obstacles for charging up cars as they have for getting efficient appliances and good insulation: They get the benefits of the investments in charging facilities or insulation. But the landlords spend to provide the equipment. What about them?

Well, many don't have cars, and many others use mass transit. I would think that this will encourage the use of carsharing (like zipcar.com ). The remaining will need public infrastructure: outlets in parking meters, parking garages, or gas stations. Fortunately, that's a small %.

Engineer-Poet said at January 12, 2010 5:21 AM:

Nick, isn't that 80 MW container-ship engine a Wartsila-Sulzer direct-drive diesel, not electric?

All of this worrying about how people without covered parking will plug in is just silly.  There's been no need for wiring to every parking space outside of places like Saskatchewan and Alaska, so those are the only place it exists.  As the need expands, so will the supply; if the PHEVs come, they will build it.  Armored cables and credit-card terminals are just details.

Nick G said at January 12, 2010 9:27 AM:


Yes, you're right, I was thinking of the auxiliary engines/generators: 40,000 hp from
five Caterpillar 8M32's . Here's a ferry that is completely powered by the same generators.

I believe the US DOD is going to electrical drive trains for tanks, and considering them for aircraft carriers - it's
an electrical future.

Nick G said at January 12, 2010 9:43 AM:

And, of course, all submarines have electrical drivetrains, whether the generators are diesel or nuclear powered.

Hard to imagine an environment which demands higher levels of reliability (or lower levels of noise).

No, there's absolutely no question that a large, engineering-driven company like GM can be absolutely sure of both the feasibility and costs of an EV. The battery chemistries; the details of the optimization of the generator behavior; and electrical accessory devices are pretty much the only new things. The optimization of the generator behavior is software: that kind of work can certainly be complex, challenging and complex, but in the end it's a finite optimization task, and the R&D will be amortized over a very large volume of vehicles (unless, of course, Better Place gets there first, or we discover abiotic oil). Similarly, electrical accessory devices will be cheaper, simpler and more reliable than hydraulics and drive belts.

So, that leaves the batteries, which brings us back to where we started. Lead-acid would work, but they're not sexy. Li-ion will get us there - the only question is how fast?

Nick G said at January 12, 2010 10:34 AM:

oops - only some nuclear subs have electrical drivetrains (French and Chinese) - others generate steam.

Nick G said at January 12, 2010 1:16 PM:


Somehow I dont think the all electic car matters all that much. I agree with what Randall said that the Prius is the target to beat. Guess what? I can guarantee that it will be beaten in 4-5 years....by the next generation Prius!

I can't really disagree - I also suspect the plug-in Prius will be very hard to beat cost-wise.

OTOH, we have recognized the true costs of oil (which are much higher than the out-of-pocket costs - how many trillions of value were lost in the recent recession, partially caused by oil?) and eventual economies of scale, via public policy with a $7,500 tax credit, and the Volt is likely to be competitive right out of the gate.

Battery costs will continue to fall, market prices for oil will rise, and reasonably soon ErEVs will be the sweet spot even with no Pigovian public policy.

Nick G said at January 12, 2010 1:22 PM:

Personally, driving something that allows me to use no oil at all most of the time, and no oil all of the time should I choose some minor inconvenience, is very attractive. I see a lot of non-out-of-pocket expenses to me personally, and I'm willing to pay a premium for that (as well as for the better performance). I think many others are as well, and that will grow, as being "green" becomes more and more high-status.

Again - if everyone was just trying to minimize their out-of-pocket cost per mile, we'd all be driving $12K, 35MPG Korean sub-compacts, right?

Bruce said at January 12, 2010 3:05 PM:

Not all ...

"Korean automaker Hyundai Motor Group, which includes both the Hyundai and Kia brands, posted a combined sales gain of 42% to 33,797 vehicles."

"The only major automaker that reported an increase in sales [for the year 2009 as a whole] is Hyundai Motor Group. Its sales rose 9% to 735,127 vehicles.

"Auto sales soared 15% to 1.03 million in December after the market bottomed in November. Sales exceeded 1 million for the second time in the year, trailing only August – the month of Clunkers program.

Most of the prominent automakers, except General Motors and Chrysler, reported a sales gain during the month. Among the domestic automakers, Ford Motor recorded the best sales gain. Meanwhile, Hyundai Motor Group ruled in terms of U.S. sales gain among all automakers."

"Sales of Ford, Lincoln and Mercury branded cars rose 42%"

"General Motors (GM) reported a 6% decline in sales ..."


Which will happen first ... the first 100,000 Volt sales, or the last 100,000 GM sales.

Nick G said at January 12, 2010 3:51 PM:

Auto sales soared 15%...General Motors (GM) reported a 6% decline

Well, it didn't help GM to drop Pontiac, Saab, Hummber, etc, which accounted for about 15% of it's sales. That alone accounts for 71% of the difference between GM's sales trend and the industry overall.

Plus, there are people are waiting for GM (and Chrysler) to come out of government ownership before they buy from it again.

Also....some people are delaying their purchase until the Volt comes out! GM is advertising the heck out of the Volt, to get it's own version of the Prius halo, but of course, that means some people will want to buy the thing! GM dealers are grumbling, and the GM board is throwing money at the Volt program to get it out the door before the long-planned 11/2010 launch date, so that people will start coming into dealerships to look at it (and, of course, hopefully buy something else, because of course there will be a shortage of Volts to begin with, just like there was with...the Prius).

A slogan I'm starting to see - NPNS (No Plug No Sale).

Ted Savage said at January 14, 2011 7:59 PM:

I have to concur with nick, there are a lot of folks using carports, and it is a bigger industry then i initially thought, it seems quite a few people have decent coverings or options for them at the very least, which i learned by browsing this site Carport Place.

Joel said at February 22, 2011 1:24 AM:

agreed Ted, You have found a great carport resource but do they actually install the solar panel Carports? I wonder who installed the ones at the google headquarters?

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©