December 17, 2009
Mass Production Pluggable Hybrid Cars Decades Away?

A study by the US National Research Council finds that substantial production of plug-in hybrids lies a few decades in the future. Will battery costs really fall so slowly?

The mass-introduction of the plug-in hybrid electric car is still a few decades away, according to new analysis by the National Research Council.

The study, released on Monday, also found that the next generation of plug-in hybrids could require hundreds of billions of dollars in government subsidies to take off.

The study claims battery costs are huge and therefore the fuel saved using pluggable hybrids take too long to pay back the added costs.

GM will start selling the Chevy Volt pluggable hybrid in November 2010. Toyota will start selling a pluggable hybrid in 2011. Nissan is embracing pure electric cars in a big way and CEO Carlos Ghosn expects electric cars to make up 10% of global car sales by 2020.

How fast will battery prices drop? How high up will oil prices go due to Peak Oil? You need to know the answers to both those questions in order to accurately predict the rate of demand growth for pluggable hybrids and pure electric cars.

Share |      Randall Parker, 2009 December 17 10:45 PM  Energy Electric Cars

Mercy Vetsel said at December 18, 2009 12:30 AM:

Unless there is a major breakthrough with an entirely new battery technology, there is no reason to expect battery prices to drop substantially.

The BEV advocates are not being honest with the numbers and the price of gasoline where BEV's become economical ranges from $10 to $15 per gallon. Long before oil became that expensive, there would be massive shifts to other fuels like natural gas or coal gasification.

This sounds extreme, but Toyota, Honda and most of the major manufacturers recognize the fundamentally bad economics of building battery cars and they are just going through the motions to claim some of the absurd piles of money that the government is throwing at battery cars.

Absent a major breakthrough that drops the cost of batteries by a factor of two or three, BEV's are never going to take off.

Mercy Vetsel

K said at December 18, 2009 9:51 AM:

You can get studies with any numbers you want. And that means one or another will be exactly right. Is this the one? Since there is a big paywall I am not going to spend time or money finding out. With (considerable) patience you can read the pre-publication online.

Vetsel is right in saying that companies will stay in the game for subsidies. And the better known will also keep a presence for prestige.

Even so I suspect this report is quite mistaken and overly pessimistic.

As Randall has said for years, it is all about batteries and improving batteries. Can they be produced cheaply enough?

Public and private money is going into battery research and production at an astonishing rate. And by many companies. What is driving the private decisions? Are the fundamental cost trends, sans subsidies, favorable for mass production of BEVs within a decade?

This is similar to decoding wind costs and solar costs; injecting public money, or even the prospect of public money and mandates, produces decisions that purely private investors would shun.

IMO the Prius and other hybrids have shown that plug-ins can make sense. We will at least see a lot of them. I suspect BEVs will be high end vehicles for a while. i.e. Vehicles above $70K where the buyer is not really buying transportation and the cost of the batteries is a smaller pat of the purchase price.

JAY said at December 18, 2009 10:12 AM:

There's no such thing as peak oil.
Calm down.
Buy an SUV.
Enjoy life.

Nick G said at December 18, 2009 2:27 PM:

I'm disappointed - the NRC is really off the mark.

The manufacturer of the Volt's 16KWH battery says that the battery pack only costs $8k , and the cells only cost $350/KWH.


Here are retail costs of $350/KWH:


We can easily see why this article is so unrealistic.

Click on the link provided for the report, and look at the authors: the first author is the Committee on Assessment of Resource Needs for Fuel Cell and Hydrogen Technologies!

Why should we be surprised that they are pessimistic about electric vehicles, and recommend more research into fuel cells??


It's easy to think that large organizations are monolithic. If we've seen good work from an organization like the NAS, we assume it's other work will be reliable.

Unfortunately, large organizations are built up of many individuals and departments, with varying levels of competence and integrity, and different interests and biases. We have to review each new product with fresh eyes.

In this case, the report flunks.

Nick G said at December 18, 2009 2:36 PM:

Randall, a typo: Nissan is forecasting 10% by 2020, not 2010.

Wolf-Dog said at December 18, 2009 8:13 PM:

Batteries are expensive, but on the other hand, because electric cars have very few moving parts in comparison to gasoline cars, once there is mass production, a reasonable electric car without battery can be sold for $12,000.

For the record, it is estimated that currently the cost of the battery for the latest electric car of Nissan is $10,000 (this battery has a range of only 100 miles and a life of 100,000 miles, and it gets swapped at the battery swapping stations instead of gas stations when there is no time to charge the battery). But in 10 years, it is very likely that the range of this battery can be increased to 250 miles for the same price.

Now because the cost of electricity per mile is much lower than gasoline, if the electric company directly owns the battery and rents the battery to the driver, the cost can be as follows:
Assuming that the battery will cost $10,000 and that it will have a life of 100,000 miles, this means that the battery alone will cost $0.1 per mile. Now the cost of electricity for an electric car is about $0.03 to $0.04 per mile. Thus if we combine the cost of using the battery plus the cost of electricity, we see that the cost of driving an electric car would be $0.14 per mile, but the car can be purchased for only $12,000, at a price lower than most gasoline based cars.

On the other hand, a gasoline car with 30 miles per gallon and fuel selling for $3 per gallon would cost $0.1 per mile to drive, which is only 40 % cheaper than the corresponding pure electric car.

But this is with the assumption that the battery will stay expensive for a long time.

Now if the life of the battery is raised to 200,000 miles at the same cost of $10,000 per battery, then the cost of the battery suddenly becomes $0.05 per mile, and with the cost of electricity $0.04 per mile, this becomes a total of $0.09 per mile (battery per mile plus electricity per mile), which is competitive with a gasoline car which costs $0.1 to drive (assuming that it gets 30 miles of gallon at $3 per gallon.)

Also, note that electric cars have very few moving parts and very few components (with the battery replaced every 10 years) and therefore would last much longer than gasoline cars. In addition, gasoline cars do NOT require the major annual maintenance that gasoline cars require, since only the brakes and tires need to be checked frequently, and there is no air filter, radiator fluid, exhaust pipe, transmission oil, engine oil filter. Thus the hidden costs of operating an electric car are much lower than the corresponding gasoline car.

Thus if we consider the longer life of the electric car and its lower maintenance costs, plus the lower price of an electric car (when mass produced WITHOUT battery), it follows that if it is true that a battery with a life of 100,000 miles can be produced for $10,000, then the cost of using such an electric car is not too bad. Once the gasoline is over $4 per gallon, then it becomes even more competitive even with the current expensive batteries.

But the real benefit of electric cars would be if the batteries are made in the US, otherwise we will graduate from foreign oil dependency to foreign battery dependency.

(So far we are NOT yet discussing the range of the battery, whether it will be 100 miles per charge or 300 miles per charge, although the current Nissan electric cars already have 100 mile ranges per charge).

Wolf-Dog said at December 19, 2009 8:02 AM:

Here is some more progress in the battery front:

Panasonic has improved lithium-ion batteries (for laptops) with higher power density. But the same small modules are assembled to make electric car batteries by Tesla. Similar technology can be used to build large batteries directly. With improving mass production technologies, the prices should decline a lot more within five years because there is so much money to be made here that a lot of companies are working round the clock to design better batteries. If it were just for laptops, the batteries would have been of limited importance and the market would have stayed very small.

Model NCR18650
(Current model) NCR18650A
(new high-capacity model)
Capacity 2.9 Ah 3.1 Ah
Volumetric energy density 620 Wh/L 675 Wh/L
Mass approx. 44 g approx. 44.5 g
Voltage 3.6 V 3.6 V
Charging voltage 4.2 V 4.2 V
Capacity 10.4 Wh 11.2 Wh

December 19, 2009 in Brief | Permalink | Comments (1) | TrackBack (0)
Comments:Now this is (1000/44.5)*11.2Wh = 252 Wh/kg
Tesla use 8000 of the 18650 cells to get a 70kWh battery for their 300 miles version of the Model S using about 0.233 kWh per mile.(1) If Tesla used this new cell from Panasonic they could make a (8000*11.2Wh=) 89.6 kWh battery and get a (89.6/0.233) = 384 miles range for the Model S.

Nick G said at December 19, 2009 2:25 PM:

That's a good article about the Nissan Leaf.

$10K for a 24KWH battery pack gives about $400/KWH: that's a little less than for the Volt battery pack. That suggests a cost for the Volt battery pack of about $6,700, or less than half that assumed by the NRC report.

Wolf-Dog said at December 19, 2009 10:40 PM:

So far it has been claimed that the $10,000 battery of Nissan will yield a range of 100 miles, which is much more than the 40 miles Chevrolet Volt is supposed to have. It is possible that Nissan understated the cost of the battery in anticipation of the lower prices when the car is really mass-produced by 2011. But the fact is that Israel and Denmark have already committed themselves to totally populate all streets with charging pods and to build a lot of battery swapping stations that will replace almost all their gas stations after 2020. This is a very serious commitment.

Here are the videos of Project Better Place and Shai Agassi:

Nick G said at December 20, 2009 9:17 AM:

So far it has been claimed that the $10,000 battery of Nissan will yield a range of 100 miles, which is much more than the 40 miles Chevrolet Volt is supposed to have.

Nissan's battery lease allows them to not worry about battery life, so they can discharge the battery much more deeply.

It is possible that Nissan understated the cost of the battery in anticipation of the lower prices when the car is really mass-produced by 2011.

Possibly, but the NRC study is also talking about projected prices in 2010-11. Between the two, I believe Nissan - their figures are consistent with those of GM and Tesla, as well as the retail prices I showed above. I think the NRC study is just making things up (as seems to be common...).

Bruce said at December 20, 2009 9:55 AM:

For those of you who are trashing the NRC study, what real progress has been made in the last 10 years on cost that makes you think ?

The NRC says 18,000 more for a plug-in hybrid. Plug-In hybrids require an electric motor, batteries and an IC engine. Sounds like a reasonable estimate to me.

As I said in an earlier thread, Tesla is betting on a $12,000 battery pack in 7 years for their car.

And do try and remember, the only time an electric car is significantly cleaner than an IC car is when the electricity is generated by hydro or nuclear. If it is coal, forget it.

Nick G said at December 20, 2009 11:00 AM:

The NRC says 18,000 more for a plug-in hybrid. Plug-In hybrids require an electric motor, batteries and an IC engine. Sounds like a reasonable estimate to me.

A good comparison is a Prius. A Prius has an electric motor, batteries and an IC engine, and only costs $24K. Adding a plug is trivial, so the additional cost of a plug-in Prius is a larger battery. I've given plenty of information above showing that the additional battery cost is less than $8K.

Tesla is betting on a $12,000 battery pack in 7 years for their car.

That's a 52KWH battery pack, so $12K would be $230/KWH. That should make the 16KWH battery pack discussed by the NRC report cost only about $4K. That makes the NRC estimate even more clearly unrealistic.

Bruce said at December 20, 2009 11:32 AM:

From what I've read, the assembly of the battery pack is hugely expensive.

"The Tesla's battery pack is rated at 53 kWh which is enough to power an average U.S. home for 2 days. It is assembled from 6831 individual type 18650 batteries, which are cylindrical cells with about twice the volume of an AA-size battery. They are the same cells used inside laptop batteries. These individual cells are connected together and packaged with a charge controller to monitor and level the charge of the individual cells. Just the wholesale cost for the batteries for that pack would be about $39K assuming a cost of $.74/Wh for Li-ion cells, which I have confirmed with a few Chinese battery suppliers as the typical wholesale pricing for Li-ion cells when purchased in volume.

In addition to the raw cost of the batteries, there is the labor cost of assembling the cells and the cost of the charge controller and housing. It also has a sophisticated arrangement of sensors, microprocessors, and its own liquid cooling system. When you add the assembly labor, controller, and packaging to hold 750 lbs of batteries, my estimate is that this pack costs somewhere in the neighborhood of $50K to manufacture."

"Provided Li-ion technology can maintain an 8% annual level of improvement, it will take about 18 years for the wholesale cost of the battery pack to approach $12K."


Wolf-Dog said at December 20, 2009 2:24 PM:

The Tesla method of assembling electric car batteries from the off-the-shelf laptop batteries, is obviously a temporary solution that was chosen in order to be the first successful electric car manufacturer at a time no electric car batteries were available. The current Tesla method is artificially expensive because currently the market for laptop batteries is more or less stabilized since less than 10 % of the price of a laptop computer is due to its battery, not to mention the cost of assembling the battery and building the complex cooling system, which will not be needed when specialized spebatteries are built only for electric cars. This is why the Nissan electric car batteries will cost $10,000 for a range of 100 miles in 2011. By 2020 it is safe to assume that a $5,000 battery can be manufactured with a range of 250 miles.

Nick G said at December 20, 2009 2:44 PM:

Bruce, that article doesn't provide any sources. He cites un-named Chinese sources, but they and his opinions disagree with my sources given above, including GM's supplier, Nissan, and a retail source of li-ion batteries at $350/KWH. Could you find his sources??

He suggests that Tesla doesn't discuss their costs, but Tesla stated about 2 years ago on their website that their battery cells cost, at that time, $20K (about$400/KWH).

I do agree that Tesla has chosen a relatively expensive path, due to their wanting an unusually large battery pack (and therefore needing higher power-density), and freezing their design earlier than others. Each li-ion cell requires electronic controls - the large-format cells everyone else uses are substantially less expensive to manufacture as well as to assemble. They're using 1st-gen cobalt li-ion: cobalt is expensive, and the requisite thermal-runaway protections are also expensive.

I see Wolf-Dog has posted as well - what he said is consistent with the above.

Nick G said at December 20, 2009 2:49 PM:

Oops - I noticed an edit: "(and therefore needing higher power-density)" should be "(and therefore needing higher energy-density)" . The italics are just to emphasize the correction.

Bruce said at December 20, 2009 3:35 PM:

I think Nissan is predicting increased range at the same cost by using Nickel Manganese Cobalt. Not twice the range at half the cost.

The Leaf will be 35-42,000 without government subsidies.

As for 10,000 for the Leaf's batteries ... maybe. I've read 450-1000 per kWh. Thats 24,000 at the the top end of the estimate. It isn't in Nissan's best interest to admit to such a high cost.

Nick G said at December 20, 2009 3:54 PM:

Bruce, do you have any sources for these numbers? My sources don't agree, and you've seen them.

I see no reason to doubt Nissan - ultimately they're going to have to sell these things. People associated with US car makers (GM in particular) used to say similar things about the Prius (too expensive to make, will never make a profit, Toyota is misleading everyone about their costs, etc, etc), and they've been firmly proved wrong.

Randall Parker said at December 20, 2009 4:09 PM:


I do not see a reason to doubt Nick G's sources on costs. You need better sources if you are going to disagree.


Going from hybrid to plug-in hybrid involves more than just a bigger battery. Plug-in requires:

- Connector and circuity for converting 110V/220V wall electric power into battery DC power.
- A stronger electric motor that can power the car to a much higher MPH. Last I checked the Prius electric motor couldn't even go to 50 MPH.
- Higher discharge rate on the battery.
- Of course a bigger and longer lasting battery.

So HEV to PHEV requires a big cost jump. Going from PHEV to EV involves cost changes in both directions:

- No ICE and fuel tank and transmission needed.
- But much bigger battery needed.

I am puzzled why I see $30k for the Nissan Leaf without battery. Why so high?

John Moore said at December 20, 2009 5:48 PM:

Why a hybrid at all? If the batteries get better, a pure EV seems to make a lot more sense.

What would the battery requirements be for a pure EV that can take a family of 4 and their luggage for 250 mi, in air conditioned comfort with power steering and brakes?

How is that X+oxygen battery tech coming? It seems to be the only one with potential for a truly large increase in specific energy density - since it gets one side of the reactant from he air?

Bruce said at December 20, 2009 8:09 PM:

"In the next five years, he believes it may be possible to lower those costs to $500 per kilowatt-hour."

"A well equipped C-class vehilce runs in the $28,000 to $35,000 range, without the $7500 tax credit the car will be expected to enjoy. If they are factoring that in, consider $35,500 to $42,500.

Now of course it isn't likely that one will be able to just go out and buy a LEAF in 2010 for that price. You see the most expensive part of the car is its 24 kwh lithium ion battery. Though Nissan is jointly manufacturing the battery with its partner NEC, costs of production for these groundbreaking laminate cells are considerable.

Usually lithium ion cells cost anywhere from $450 to $1000 per kwh. That puts the pack alone somewhere between $10,000 and $24,000."

Wolf-Dog said at December 20, 2009 10:21 PM:

Another detail: The additional cost of sending 30,000 more troops to Afghanistan will be $30 billion per year:

If Obama had given $30 billion per year just for battery research, within a few years we would have far better batteries good enough to make US independent from foreign oil.

What we need is to organize a movement to write letters to the US government to ask them for accelerated funding of battery research and electric vehicle infrastructure (such as putting charging pods in every street.)

Nick G said at December 20, 2009 10:34 PM:


Connector and circuity for converting 110V/220V wall electric power into battery DC power.

This is a trivial cost. AC->DC power supplies are relatively inexpensive (though installation of 220V or 440V in your garage might cost a few $100s, due to the need for an electrician. Also, rectification and inversion are already needed by a hybrid's power train, so you may literally just need a plug.

A stronger electric motor that can power the car to a much higher MPH. Last I checked the Prius electric motor couldn't even go to 50 MPH.

Absolutely. Conversely, the ICE can be made much smaller. For instance, the 2010 Prius has a 1.8L engine, while the Volt will have a 1.4 (and it would be much smaller if they weren't saving money by using a large-production, off-the-shelf engine). The next-gen Volt will have a custom generator/ICE. Further, all other ICE supporting components can also be downsized. FWIW, the 2011 plug-in Prius will have a 62 MPH electric top speed.

Higher discharge rate on the battery. Of course a bigger and longer lasting battery.

Those two go together: a larger battery will give the power needed without straining the battery.

So, really, it's just the battery.

I am puzzled why I see $30k for the Nissan Leaf without battery.

Keep in mind that pricing is currently highly speculative. If true, it's because Nissan wants to capture the $7,500 credit, and thinks that there's sufficient early adopter interest to support it.


Note that the $500/KWH future figure is from a financial analyst. He's not a primary source.

Read the allcarselectric blog post carefuly: it's pure speculation, working back from very rough early figures from Nissan. We have the best possible source within Nissan telling us the figure is $400/KWH.

Bruce said at December 20, 2009 11:16 PM:

Nissan is planning to LEASE the battery pack. Therefore I am quite suspicious of its real cost and lifetime.

The new plug-in Prius is going to be 5-8000$ more for a 4kWh battery pack - a jump from 1.6kWh

I don't believe the Nissan Leaf's pack can be 6 times as big as the Prius and only cost a little bit more.

Nick G said at December 20, 2009 11:43 PM:

Nissan is planning to LEASE the battery pack. Therefore I am quite suspicious of its real cost

Buyers discount operational savings very heavily, and the batteries will have a large residual value (75% of capacity remaining). Leasing just makes sense.

and lifetime

We can certainly expect it to have a much shorter lifetime than the Volt and Prius batteries - they're quite upfront about that. Given the rapid decline in battery cost, that's sensible.

The new plug-in Prius is going to be 5-8000$

No, all we have is a ceiling of $10K. The rest of the article is pure speculation.

Bruce said at December 21, 2009 10:06 AM:

Nick, I read your reference. 350$/kWh just for the cells, and no committment on price for all the other things involved. And maybe half the cost in 10 years for the same capacity.

Until you come up with a better reference, I'll go with 1000$/kWh for the whole battery pack.

kurt9 said at December 21, 2009 11:07 AM:

I think the article is correct. Batteries are a lousy technology and they do not seem to be improving. Battery performance decreases over time. It takes longer to charge them and the charge they hold decreases. They are expensive to replace as well. I write this as I recharge my MP3 player for a flight later today. It used to take about 45 minutes to recharge my player. Now it takes closer to two hours. I will probably buy a new MP3 player upon my return in January. The same story is true for cell phones. These batteries degrade over time as well. I've had my current laptop for 2 years and have rarely run it off of the battery.

I think battery technology is much like solar power. There is much hype and promise, but very little delivery on that hype and promise.

The future is synthetic hydrocarbon fuel produced from nuclear power. Much of the West has its head up its ass about nuclear power, but the Asians are pushing ahead with it. I expect China to be the first country to start producing synthetic hydrocarbon fuel using nuclear power. Everyone else will follow.

Nick G said at December 21, 2009 3:17 PM:

no committment on price for all the other things involved.

I'm not sure what you mean. CPI says that the battery pack is $8K, or $500/KWH for 16KWH. That covers all battery costs, right? That works out to $5,600 for the cells, and $2,400 for the power electronics that provide depth of charge and temp management, as well as the cooling system.

Now, Nissan says that their 24KWH pack will cost $10K: if we subtract $2,400 for the pack overhead, that gives us $7,600 for the cells. $7,600 divided by 24 gives us $320/KWH, which is pretty close to the CPI/Volt cell cost.

Further, I gave you a retail source where you can buy cells for $350/KWH.

So, cost estimates from the two biggest suppliers, and a retail source. How can references get better??

Nick G said at December 21, 2009 3:20 PM:


You're talking about older cobalt based li-ion batteries, with no temp or discharge management. Even Tesla is doing much better than a laptop, using older batteries, with the right battery management.

The future is synthetic hydrocarbon fuel

That sounds great. Do you have any detailed information? Costs?

Nick G said at December 21, 2009 3:33 PM:

John Moore,

Why a hybrid at all? If the batteries get better, a pure EV seems to make a lot more sense.

The batteries would have to get a lot better in the absence of infrastructure. Without lots of places to charge at high power, you need a very large battery. That will be too expensive for a while.

Some people are OK with short ranges, but I think that's a pretty small niche.

What would the battery requirements be for a pure EV that can take a family of 4 and their luggage for 250 mi, in air conditioned comfort with power steering and brakes?

The Volt gets 4 miles per KWH. The Nissan leaf probably gets about 75 mile range with all accessories. If we take the Nissan approach and use about 75% of the battery pack, then you need 250 / 4 / .75 / = 83KWH. At $350/KWH that's about $29K for the cells - add $2,400 for the battery pack and you're close to $32K just for the battery pack. That means the car has to be at around $50K.

Bruce said at December 21, 2009 3:44 PM:

Nick, where is the 500$/kWh quote from CPI? I can't find it in your reference. Wishful thinking on your part?

Nick G said at December 21, 2009 6:34 PM:


Either the article has been edited, or I missed something. Anyway, here's an article which clarifies: costs of $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."

Randall Parker said at December 21, 2009 8:52 PM:

Nick G, Imagine an SUV that uses 500 wh/mile and imagine you would want a 500 mile range on it so you could use it on long trips without a big battery swapping infrastructure. So you'd need 250 kwh of battery capacity. At your more optimistic $350/kwh that's be over $90k just for the batteries. But that's assuming full discharge. If you do only 75% discharge then your battery costs go north of $120k. You'd need $35/kwh to make the batteries affordable.

By contrast, imagine you live 12 miles from work and can recharge every night and drive a compact electric car that uses 250 wh/mile. Then even if you gave yourself 4 miles extra range and 75% discharge you'd need only 5.3 kwh of batteries for your daily commute. So only about $1855 for the batteries.

Of course, using Bruce's higher number for battery costs that long range electric SUV would cost more than a Ferrari.

Peak Oil will mean that shorter trips cost far less per mile than long trips.

Bruce said at December 22, 2009 8:35 AM:

Thanks Nick. 1000$/kWh it is for AT batteries (cars) versus 350$/kWh for CE (consumer electronics).

Therefore if the Nissan Leaf has 24kWh = 24,000 for the batteries.

Engineer-Poet said at December 22, 2009 10:28 AM:

I was wondering when someone would note the high residual value of the battery pack (esp. one using cobalt cells, which has a very high scrap value); this will decrease the per-mile cost and lease payment considerably.  Kudos to Nick G for hitting the bullseye on that one.  A second life for such packs would be in utility load-levelling or as an energy buffer so service stations can provide rapid charges without overwhelming the grid.

I think Mercy Vetsel is off the rails here.  Petroleum prices of $10/gallon before taxes will collapse the economy (a large fraction of petroleum goes for industry and heating, and natural gas would follow suit), and the USA's politics are so insane there is no chance of pushing a changeover with taxes before the oil exporters suck the capital overseas.

Last, kurt 9 is touting the "Green Freedom" hydrocarbons-from-nuclear boondoggle.  If you look at the capital costs and production rates from the proposal, you'll see that it costs much more than electric propulsion and requires several times as many reactors.  I think this is another red herring from petroleum interests, because a system which uses hydrocarbon fuels will leave a strong market for oil while an electrified system has none.

Nick G said at December 22, 2009 11:02 AM:

Bruce, it's $500/KWH for the Volt battery. The $1000/KWH is for "effective capacity" for the Volt: all batteries are only partly used, to extend their lives. The Volt battery is only 50% utilized, the Nissan battery is roughloy 75% utilized. The CPI CEO is simply trying to explain all the parameters of the Volt battery - he does it in a little bit of a backwards, confusing way.

It's a little like electrical generation - for instance, many gas plants only run 20% of the time. But, prices for generating capacity capex are normally for "nameplate" capacity.

So, the bottom line: the Volt battery is $8K, the Nissan battery is $10K.

Nick G said at December 22, 2009 3:23 PM:

Randall, some thoughts.

I agree - a BEV with a long range is going to be prohibitively expensive for quite a while (that's the one point on which Mercy Vetsel is correct - of course, this article is about PHEVs, not EVs).

A PHEV makes much more sense. The Volt will give an 80 mile range with one charge in the middle, which accomodates 80% of VMT even now.

As far as long-range operation goes....that's not so simple. We can expect quite a bit of evolution of vehicles - they're optimized for cheap liquid fuel, not electric operation. For instance, electric operation makes aerodynamics much more important than weight, especially for highway operation.

If you designed an SUV that had the stylying of an Aptera, you could create a 3 ton vehicle which only used .125KWH/mile. Onboard generators may get much more efficient. Even if liquid fuel goes to $10/gallon (which, I agree with E-P, is very unlikely), and with current generators that only produce 10KWH/gallon, that would only be $.125 per mile, roughly what SUV's get now.

A Volt gen 1 would only cost about $.20/mile for long-distance travel, even at $10/gallon (the volt gets .2KWH/mile battery/generator to wheel). The gen-2 will do better.

For highway operation you might only need infrastructure on highways, which greatly reduces the investment, and brings closer the day of all-electric operation.

Bruce said at December 22, 2009 3:29 PM:

The second life for battery packs is extreme wishful thinking.

And Nick, its 1000$/kWh for cars. Not 500$/kWh.

As for relying on GM's ability to determine how much it costs to build something .... ha ha ha ha ha!!!

Nick G said at December 22, 2009 3:38 PM:

edit: "A PHEV makes much more sense. The Volt will give an 80 mile range with one charge in the middle, which accomodates 80% of VMT even now." should be "A PHEV makes much more sense. The Volt's 40 mile range accomodates 80% of VMT even now - a 80 mile range using one charge in the middle would probably capture close 85+% of VMT."

Nick G said at December 22, 2009 3:47 PM:


The second life for battery packs is extreme wishful thinking.

Not at all - it's based on standard specifications for battery life: when 25% of discharge capacity is gone, you replace it. The decline rate after that is slow, and the packs could be easily used for something else.

its 1000$/kWh for cars. Not 500$/kWh.

No, it's really not. The $1000/KWH figure above was for "available capacity" of 8 KWH, giving a total pack cost of $8K. If you want to specify cost in the normal way, which is by total cell capacity of 16KWH, then it's $500/KWH.

"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."

As for relying on GM's ability to determine how much it costs to build something .... ha ha ha ha ha!!!

We're not. We're relying on CPI, which is an affiliate of LG (which is actually manufacturing the battery cells). LG is an enormous company - you may have one of their cell phones or appliances - and very reputable as an engineering company.

Nick G said at December 22, 2009 3:55 PM:


I agree with your comments.

Now, about "Green Freedom". I take it you've looked at the capital costs and production rates from the proposal. I'm curious: does it use nuclear plant heat, or power? If it's power, let's assume we have wind power generating wholesale power at, say, 6 cents per KWH: what would the synthetic fuel cost?

Liquid fuel will be convenient for some uses for quite some time. I'd be curious to know what the upper bound would be on the cost of liquid fuel.

Engineer-Poet said at December 22, 2009 7:37 PM:

"Green Freedom" is essentially H2CAR with a nuclear prime mover.  I dissected H2CAR in this post at The Oil Drum.  Pretty much all the downstream calculations apply to both, and given that the thermal efficiency of supercritical CO2 turbines is close to the efficiency of thermochemical hydrogen production, even that part is close enough for the conclusions to apply without change.

Randall Parker said at December 22, 2009 7:59 PM:


Regards a second life for batteries: Well, if they last 10 years in cars then the second life doesn't begin for 10 years. At that point what percentage of their original market value still exists? I see that as a function of a few things:

- Demand for battery storage for stationary purposes in 2020.
- Price declines for making new lithium batteries. Why buy used if new is cheaper?
- Prices for non-lithium much cheaper stationary storage batteries. How much less is a zinc-based or lead-based battery today? Lithium has to fall at least to that level in the aftermarket once the 10 years of car battery usage is up.

My guess is that the residual value will be pretty low. But if you can tell us what the dollars/kwh/year (adjust batteries for their lifetimes) of competing batteries are today that'd give us an idea.


Since 71% of oil gets used for transportation and only 23% of oil gets used for industry my expectation is that expensive oil will hit economy at least 3 times as hard via transportation cost rises than by industry uses. You might argue in response that it is easy to shift to 2-3 times more efficient cars.

I'd like to know a lot more about how industry uses oil. How substituteable is it?

Nick G said at December 23, 2009 9:26 AM:


The residual value of batteries came up in connection with Nissan's Leaf lease (try saying that 5 times in a row), which doesn't have the 10 year requirement. GM has pretty much designed the battery to last for the vehicle's lifetime.

2nd-gen li-ion has roughly 10x the cycle life of lead-acid, so it has roughly 5-10x the value.

Residual value declines, but so does replacement cost.

The economic impact of oil: 1st is the tax-like effect of transferring income to oil exporters. 2nd is the effect on capex of changing oil prices: consumers and business delay capex until they have greater clarity on the importance of oil prices (Do we need to trade-in our SUV for a small sedan? Let's wait and see...). The experience in the US is with oil shocks - PO pessimists fail to realize that to a large extent it's not the high price per se, but the change in prices that causes problems. This means that the problem is temporary, and not due to a shortage of BTU's, but to uncertainty.

Businesses need their material inputs more, but have greater economies of scale, so they respond to price signals better. Hydrocarbon feedstock is the big industrial oil consumer: hydrocarbons can come from coal, nat gas and biomass. The switch to something new takes R&D and capex (and thus has substantial response time-lags), but some efficiencies and substitutions are easy, like thinner pop bottles.

Nick G said at December 23, 2009 10:07 AM:

Here are some DOE overviews, but not specifically for industrial consumption: and

I'm realizing that my intuition isn't very good about industrial uses - I think I'm exaggerating the importance of plastics. I'd like to see a nice explanation of industrial uses of LPG's (propane, butane, ethane, etc) and fuel oil.

Randall Parker said at December 23, 2009 6:51 PM:

Nick G,

I just happened to run across a very good source on expected future residual value of car batteries: An article by Michael Kanellos from his talking with Ford executives about future improvements in Ford car efficiency. At most 30% expected residual value:

As battery prices go down, the cars will become more attractive. Potentially, up to 30 percent of the cost can be covered by re-selling slightly depleted batteries back to utilities.
Nick G said at December 26, 2009 4:26 PM:

That's a good data point, but I'd like to see more supporting detail. For instance, I wonder of that's 30% of the cell cost, or of the battery pack?

In this context, one might assume they were talking about overall battery cost, which includes the pack. 30% of the battery pack cost would be great. 30% of the Nissan battery pack would be $3,000: that's 40% of the $7,600 cell cost (the battery pack wouldn't be very useful to an utility). That's about as good as one could hope for.

Even 30% of the cells wouldn't be bad. The pack control costs will fall much faster than cell costs.

Engineer-Poet said at December 27, 2009 9:57 AM:

I'm not sure how substitutable industrial oil use is.  That's what worries me.  Methane and propane for heat are much more easily substituted than ethylene and propylene for chemical feedstock, but if you look at production statistics ethane and ethylene are typically counted together.  I have no idea how easy it is to desaturate hydrocarbons for use as e.g. monomers.

The mass-production plug-in car should have been yesterday.  Suppose for a moment that we can make Firefly 3D² cells for the same per-kWh cost as current mass-market deep-cycle batteries (shouldn't be difficult with a 2/3 savings on lead).  That's about $60/kWh, or $960 for a Volt-sized pack.  If you can get a 3-year useful life out of them in traction use at 1 cycle/day to 50% DoD,the residual value is $15/kWH and your discount rate is 14%/year, the total battery cost is:

$(60 - (15/1.14^3))/kWh / (1095 * 0.5) = 9.1¢/kWH.  If electricity costs 10¢/kWh and the battery is 80% efficient, the overall cost is 21.6¢/kWh at the terminals.  That's equivalent to gasoline at around $1.50/gallon.  It's not an overwhelming advantage but it's viable today and in any regime we can expect except a dumping of oil stocks due to economic collapse.

The Optima Yellow Top cycle-life graph is no longer on the Commuter Cars web site, but I'd bet that 1100 cycles to 50% is very conservative for a Firefly-type cell.  At 5 years and 1825 cycles to 75%, the cost falls to 3.8¢/kWH for the battery, 16.3¢/kWh at the terminals.

Randall Parker said at December 27, 2009 2:54 PM:


Looked at as an investment that 30% recoverable after, say, 10 years does not look like a good investment. In 10 years time surely a cheap way to make stationary battery storage that is less than 30% of current prices for Li car battery storage will be developed. If that future zinc air, lead carbon, or other battery costs only 10% of current car Li battery costs then suddenly recoverable cost goes below 10%.


Chemical feedstock: But what portion are they of industrial uses of oil? What the heck are the industrial uses of oil? Surely running some generators. But what else? Is the bulk of it chemical feedstocks?

Plant fibers can substitute for some synthetic fibers, at least for most clothing. Steel, aluminum, wood, and other materials can substitute for some plastics. What industrial uses are not easily substituted?

Nick G said at December 27, 2009 5:12 PM:


The Volt's business plan doesn't rely on battery residual value - the battery is really meant to last the life of the car. It's Nissan that's thinking that way, perhaps 5 years out.

Regarding industrial uses of oil: if there are some feedstocks that are difficult to produce from biomass, it's likely only 5% of US oil consumption. Other uses will go away first, and we'll still be producing much more oil than that 100 years from now. Heck, the world could produce 10M bpd for 100's of years, given how much really heavy oil/tar there is in Canada and Venezuela.

I'm not sure, but I don't think there's any theoretical reason why one could not produce very simple hydrocarbons from biomass, and build them into any compound your heart might desire.

Randall Parker said at December 28, 2009 9:25 AM:


But imagine the world 50 years from now where China and India each have economies bigger than America. The demand for plastic will be enormous.

Nick G said at December 28, 2009 10:11 AM:

True, but I have a hard time worrying about it.

1) as you note, there are pretty good direct substitutes for many uses: metal, wood, plant fibers.

2) We have enormous supplies of fossil hydrocarbons in the form of coal, methane, kerogen (shale oil), heavy oil, bitumen (tar sands), peat, etc, etc. for the next 200 years. They have problems: they don't flow like conventional oil in a way that prevents peak oil, and some have low or negative E-ROI, but they're enormous and perfectly affordable for niche, high value uses like plastic feedstocks.

3) There's biomass, which pretty badly suited for conversion to liquid fuel, but good for hydrocarbon feedstock.

4) plastic is recyclable. We don't do a great job of it now, because we don't have to. If you recycle 95% of consumption (cars, for example, are 99% recycled), you only need to produce 5% as much.

sunnnv said at January 11, 2010 4:34 PM:

Randall -
re: - A stronger electric motor that can power the car to a much higher MPH. Last I checked the Prius electric motor couldn't even go to 50 MPH

The reason the Prius electric motor isn't used alone over about 35/42 mph (depending on the "authority") is the Power Split Device and Toyota's firmware management of Motor-Generator-1 electric motor speed (limited to +/- 6500 rpm).
The 67 h.p. main electric motor (MG2) isn't that much less powerful than the 76 h.p. Internal Combustion Engine, and there is an existence proof that it can drive 70 mph (see "tweak" below).

n.b. there are TWO electric motors, the smaller one - MG1 (about 12 h.p. IIRC) is the counter-torque motor which enables the planetary gear "transmission" to be clutch-less, starts the ICE, and does some of the regenerative braking.

A good explanation of the PSD, complete with animation you can control.

more info and cutaway pic of the real thing:

These guys tweak the firmware to allow 70 mph "all-electric" - hmmm, spin the ICE without injecting fuel ???:

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