July 30, 2009

Is there enough wind resource to provide all of our electricity?


A peer-reviewed study in the Proceedings of the National Academy of Sciences indicates that "a network of land-based 2.5-megawatt (MW) turbines restricted to nonforested, ice-free, nonurban areas operating at as little as 20%of their rated capacity could supply >40 times current worldwide consumption of electricity, >5 times total global use of energy in all forms.

Resources in the contiguous United States, specifically in the central plain states, could accommodate as much as 16 times total current demand for electricity in the United States. "

This study doesn't address changes to the grid that would be needed to supply all of our electricity from wind and solar:

"...Wind power accounted for 42% of all new electrical capacity added to the United States electrical system in 2008 although wind continues to account for a relatively small fraction of
the total electricity-generating capacity [25.4 gigawatts (GW) of a total of 1,075 GW] ...Short et al. , using the National Renewable Energy Laboratory’sWinDs model, concluded that wind could account for as much as 25% of U.S. electricity by 2050 (corresponding to an installed wind capacity of 300 GW). "

But 25% is a good start.

See here the full study in PDF form.

July 29, 2009

Should we expect climate scientists to cut their carbon footprint?

I saw an article by someone who talked to climate scientists, and found they were planning to take long plane trips. Trips like that create a lot of CO2 - doesn't that say they aren't serious about climate change?

No. This kind of analysis is misleading.

First, air travel is a relatively small contributor to CO2, and that the marginal cost of CO2 reductions from other sources is very likely much larger. Air travel isn't really the place to start. This is a good example of why public policy is better than random, ill-considered individual action. Better is a society-wide program like carbon taxes and/or cap-and-trade, which unleash the power of markets to find the easiest and cheapest ways to cut CO2 emissions.

Markets are nice and simple, in many ways. Unfortunately, institutional resistance (primarily the car makers, but also the oil & gas industry) has killed any chance of a gas tax, and is working hard (joined by other fossil fuel producers, primarily coal) on killing a cap-and-trade market approach, so all that's left is regulation, such as CAFE and the efforts of CARB.

Second, people don't want to be the only ones doing making sacrifices: they know that their individual contribution is tiny, and their personal sacrifice is very large to themselves. They want uniform rules so that everyone is sacrificing.

Sports give us some good analogies: no player is going to wear protective gear that gives them a disadvantage, but they are likely to be very much in favor of uniform rules that require all players to wear the gear.

But, if you talk the talk, Shouldn't you walk the walk, or be a hypocrite?

Sure. If a climate scientist says that everyone should take the individual initiative to stop flying, and then goes ahead and flies, he's a hypocrite.

But...is that the case? Do scientists actually say that? I suspect not. I think if you look at their public statements on What Should Be Done (if you can find any - many confine themselves to the science), I think you'll find that they recommend changes in public policy, and when they talk about individuals they mention relatively minor personal changes like CFLs and electronic thermostats: things that actually save people money, or are minor sacrifices.

We don't want to lump together everyone who deals with climate change. Is the average climatologist out there being an activist? Mostly not.

Scientists....activists..mostly different. Sure, there's the occasional Hansen, but they're mostly different groups.

Doesn't this suggest that human nature isn't conducive to solving climate change via billions of people deciding to restrain their fossil fuels consumption?

Sure. Individual actions can help a bit, but it's very clear that public policy ("rules of the game") is the big lever.

If a police chief were to advocate for a law against drunk driving, and then were to commit the currently legal act of drunk driving, we wouldn't think much of him.

Sure. I'd have to say that I'd agree that climate scientists probably are being at least a little hypocritical. Part of it is probably external influences, like pressure to go to meetings for professional reasons, and family desire for vacations. And, partly....they're human.

But...does that really say anything about climate science? Of course not. It does say something about the difficulty of dealing with climate change. I'm not all that optimistic, and I would say that we really need to develop cheaper low-CO2 tech, to make dealing with climate change less painful to implement.

Here's a comment from WCW:

hypocrisy is not interesting, and even if it were, this is not hypocrisy. Individual virtue is not a solution to collective-action problems. The way we solve collective-action problems is collectively. The fair question about hypocrisy is something like, have you organized to help stop AGW, have you voted for people who have committed to stopping AGW, and such. Any question about individual behavior is prima facie misdirection, and calls into question the motives of the questioner.

Good points.

July 27, 2009

More Volt Questions...

Are we sure consumers want this?

I don't think there's much uncertainty about consumer response. GM only decided to make the Volt because there was enormous response to the concept vehicle. EVs have been around for 100 years, and its been very clear that battery problems (cost, life, charging time, range limits, etc) have been their primary problems. They have great performance, low maintenance, low "fuel" costs, are quiet, etc, etc. Now that those problems are adequately solved with the PHEV design, new-gen li-ion, and higher gas prices, the market is wide open for PHEVs.

In an age of Peak Oil-related high fuel prices, ruinous oil-related trade deficits, oil wars, and potentially disastrous climate change (which is pretty likely to start to be priced in to market prices), isn't it nice to know that there's a cost-effective alternative, with no performance compromises, that uses only 10% as much fuel as the average US car?

Wouldn't charging at work be expensive for employer?

less than 10% of Volt owners are likely to want at-work charging (only 22% of all commuters are going to need it at all, and many of those are only going to need it for 5 or 10 miles on the way home). That's not going a large-volume problem for a little while.

As I noted before, it's likely that PHEV/EV charging will get a discount at whatever time it happens, as it's enormously useful for load following and frequency regulation (if the utility sees a spike in demand, or loses a generator, it can cut off the PHEV/EV charging in milliseconds).

If employers provided free charging, wouldn't that delay the rollout of public charging by companies like Better Place?

I think it's pretty clear that in the US the primary market niche will be for PHEVs, not EVs. Unlike gasoline, 90% of drivers have an outlet available at home, so public charging is much less important; and distances are much greater in the US. I wish Tesla, Nissan and Better Place very good luck (they'll have their market niche), but PHEVs will dominate for quite a while. And...that's just fine.

The perfect is the enemy of the good.

Why do you talk about these cars as though they were already in mass production, when they remain essentially prototypes today?

Are you concerned that the Volt might go the way of the EV-1? I think you're worrying too much. The Volt has none of the performance problems of the EV-1; PHEV/EVs are strongly supported by public policy, as shown by what the current administration says and by the planned CAFE regs; and, GM has made it pretty clear that the Volt is absolutely central to it's corporate strategy.

Electricity isn't free in either an economic or energy sense.

We don't have any real possibility of a shortage of electricity in this country; all of our electricity is domestically produced, so there's no trade deficit or security of supply problems; and PHEV/EVs supports the expansion of wind power by providing night time demand and mitigating intermittency.

Won't expensive, fossil fuel powered natural gas turbines be the primary power source for PHEVs/EVs?

No, when PHEV/EVs start to scale up, wind power is the natural source. Charging will be at night, when there's excess wind power, and charging can be dynamically matched to wind output. Smart meters will move charging to the points in time with the lowest marginal rates, which means wind and nuclear.

July 26, 2009

How much electricity does the Volt use?

We don't have all of the information we might like. On the one hand, the 1st-generation Volt gets at least 40 all-electric miles on both the EPA city and highway cycles, and does it using an effective battery capacity of 8 KWHs (50% of the nominal 16 KWHs). That gives us .2 KWH/mile, battery to wheel.

Efficiency is likely to improve with later generations - especially aerodynamics (the most important factor with an electric drivetrain, where regenerative braking greatly reduces acceleration/braking losses), but also peripheral loads. The importance of aerodynamics can be seen with the ultra-streamlined Aptera, which is expected to use only .07 KWH per mile.

On the other hand, that doesn't include the charge-discharge losses, which are usually 7-10% for li-ion, or AC-DC conversion losses, which can range from 1% to 75% (for small, cheap "wall-warts"). And, those efficiency improvements could easily go to larger vehicle size (though not to acceleration: one of the nice things about electric motors is that they get more efficient as they get larger, unlike infernal combustion engines).

You could use .25KWH/mile, if you wanted to be conservative.

July 25, 2009

But would the Chevy Volt pay for itself?

I was asked:

I only drive 10,000 miles per year, and I'd guess only half of that would be on the 40 mile electric range. Electricity is expensive, and I don't have a meter that charges less at night. Would that really pay for itself for the average driver like me?

Well, let's look at the numbers:

Annual Vehicle Miles Travelled: US total VMT is about 2.9 trillion. There are about 230M 4 wheel light vehicles, for an average of about 12,500 miles per vehicle per year. Newer vehicles are driven more, so about 13,000 miles is probably about right.

% of electric miles: For one who drives 12K miles per year, the % will be about 80%. This will seem more intuitive when one remembers that 50% of all VMT is commuting; the average commute is well below 40 miles; and even someone who commutes 50 miles will still drive 80% of the time on electric.

So, total electric miles will be about 10,000 per year for the average driver - a large minority will drive even more.

Electricity costs: the energy act of 2005 mandated that utilities make smart meters available to their customers. These programs often aren't well publicized, but they should be there - look carefully at your utility's web site. If you like, tell me your utility, and I'll see if I can find it - this wouldn't be the first time I've surprised someone by finding their utilities time-of-day metering program.

The average price of electricity in the US is about 11 cents/KWH. That gives a cost of 2.3 cents per mile ($.01165/KWH x .2KWH/mile). Any decent smart meter program should cut that in half.

If I get my employer to let me charge during the day, wouldn't that be expensive peak power?

It's likely that most employers won't find it worth their while to meter individual power outlets in the parking garage: they'll consider it a low-cost employee benefit, paying for itself with good PR. It's likely that when the cost becomes high enough to matter that the utility will have in place programs that take advantage of the load-following and frequency regulation services that EV/PHEVs will provide, and therefore charge relatively little for the KWHs to EV/PHEVs.

Will there really be enough early adopters to pay the high price of the first plug-in hybrids?

Yes. There's an enormous pent up demand, and a PHEV like the Volt doesn't have a range limitation like the EV1, and has 0-60 in 8.5 seconds (much faster than the EV1, though even the EV1 made drivers very happy) - there's no compromise at all. Also, don't forget that there's a $7,500 tax credit for buyers of vehicles like the Volt.

The EVs of a hundred years ago didn't have to contend with 21st century safety regs and product liability litigation. Voltage high enough, for example, to electrocute a first responder using the "jaws of life" to pry someone out of a wrecked EV requires some care.

Voltage problems and other safety questions for electric drivetrains were solved in the Prius long ago - that's really not a realistic concern.

July 21, 2009

Volt Battery costs, part 3

Well, at last I've found an authoritative source for the cost of the Volt's battery pack.

The CEO from CPI (the company that builds the Volt packs today) says the pack will cost $350/KWH for the cells. We saw in another article 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."


That includes the cost of the pack, with electronic controls.
For a minority of drivers, who would drive 15k electric miles per year, a Volt will pay for itself at $3.35 gas (an $8K battery over 10 years is $800 per year - a Prius uses 300 gallons to drive 15K miles, and a Volt would use 240 fewer gallons). This would include long-distance commuters (say, driving 30 miles each way and charging at work for 230 work days per year, and 10 miles per day on the other 135 days per year) and fleet drivers such as taxis whose vehicle can be used two shifts per day, and yet don't go that far and can be charged during multiple breaks (taxis typically drive 100,000 miles per year, putting 300,000 on a hybrid in just 3 years) - perhaps 10% of drivers?

Now, there are other costs: there's $.01-.02 per mile for electricity ($.01 for the average person charging at night, $.02 for during the day). But, what about the value of time? Saving 30 trips to the gas station at, say, 7.5 minutes each, is 3.75 hours. At $20/hours, that's another $75 per year. Also, maintenance costs will be less: very few oil changes, etc. Together, these roughly pay for the electricity.

The current battery might require $4-$5 gas to capture a large % of the rest of drivers. They will have to wait for the 2nd or 3rd generation of Volt, which will be less expensive, or for more expensive gasoline - whichever arrives first.

What costs are you assuming for this?

I'm assuming $24K for the Prius - I've seen news reports indicating that's the average actual US selling price (Edmunds says the base price is $22K). The same reports indicated that the average price for the US overall was $28K. Edmunds says that one high mileage competitor, the Jetta TDI, has a base price of $23,370 (I note there are a lot of options), with 33 MPG.

Now, on Volt cost analysis. I'm assuming a Prius cost, with an $8K battery added. I think it's clear that the Volt with no battery will be no more expensive to manufacture in large volume than a Prius.

Why do I think that?

Electric drive trains are cheaper than ICE (internal combustion engine) drive trains. Heck, a ten year old can build one from scratch with wire, cardboard and pliers (with really good instructions.....) - I don't think anyone can say that about an ICE. A Volt is an EV with an onboard backup generator and a lot of good programming of the electronic controls. That won't be any more expensive than a Prius, which also has dual drive trains. Heck, the Volt should be cheaper, as the auxiliary ICE support systems can be smaller.

Electric drive trains are oooooold. EV's were sold in large volumes 100 years ago, commercially, until cheap gasoline killed them. GM sold electric trucks in large, commercial volumes from 1912 - 1918. There are something like 30,000 EV conversions on the road in the US (you'd be amazed what hobbyists do). Submarines have had them for what, 80 years? Freight trains have them. The largest container ship in the world has them. There are many tens of millions of small, non-highway legal EVs in use. It's very likely that there are more electric motors in use in the world than ICEs.

EVs are easy to do. Optimizing them, as well as the batteries, to make them as competitive as possible (which is what GM and other companies are obsessing over right now) is good old fashioned engineering - no rocket science*, no tech breakthroughs. PHEVs require a bit more work to optimize the connection between the backup generator and the electric drive train, but that's good old fashioned programming.

Now, the latest batteries do represent tech breakthroughs, but that's done. All that remains is ramping up production volumes and getting prices down. Is there any question that will happen? Not really. I think one can be rationally skeptical of Tesla: it's a small company, and perhaps it will fail. But the major car companies, like GM? Not now, with a very clear US gov public policy in place, and rising gas prices to back that up.

Do we need better batteries?

No. It would be really nice to have something like the Firefly new-tech lead-acid, or the Eestor ultra-capacitor make batteries really cheap, to really make it clear that the ICE era was over.

But, it's not necessary in order for EVs and PHEVs to compete with $5 fuel with the current battery price, and for them to compete with $3 fuel in 4 years without the credit.

*I was amused to note that the CEO from CPI quoted above has a PhD in Aerospace Engineering, so he's literally a rocket scientist.

July 13, 2009

Is the Volt battery too expensive? (part 2)

A new Wired article says that electric cars cost too much, because the needed batteries cost $20,000 to $30,000.

The article is correct: Tesla tells us that their 52 KWH battery cost $20,000 wholesale a year ago (for about $400/KWH), so the article's estimate of a $20-$30K price is roughly ok. At that price, batteries are still too expensive for EVs to provide a driving range that is comparable to an ICE vehicle, at a comparable market price.

It's worth noting that this is probably not true if one includes non-market external costs - costs which are real, but not included in the market price), so for those buyers who are willing to pay for non-market costs even though they don't have to, the car pays for itself. That's a relatively small niche market, but it's real. It's also worth noting that the Tesla provides serious sportscar performance at a price that is lower than that of comparable ICE vehicles, so it's actually competitive in that niche.

But, all of the above doesn't matter, because...we don't need pure EVs. PHEVs like the Chevy Volt will eliminate 90% of liquid fuel consumption at a life-cycle cost which is comparable to, or less than, that of an ICE.

Take the Tesla battery and reduce it from 52KWH to 16 KWH, and we get a price of about $6,000. Apply the 8% annual price reduction that Tesla reports seeing in the markets (and which both NIMH and li-ion batteries have been experiencing consistently for the last 10 years) over the period 2008 to 2012 (when the Volt will get to serious volumes) and we get a price of about $4,000. That's about $300/KWH, as predicted here. "I do expect the price will come down to perhaps as low as $200 per kilowatt-hour when mass production begins in 2010 and 2011," she says." They'll use less expensive materials than 1st Gen li-ionbatteries; the larger format is much less expensive (Tesla uses about 7,000 batteries!); and they'll have very, very large production volumes relative to most 1st-gen li-ion. Large production volumes reduce costs very quickly.

The Volt's non-battery components won't cost any more than those of a Prius when manufactured in volume. Toyota is selling it a profit, at about $24K on average. It has an electrical drivetrain, and an ICE drive train. The only real difference in cost between it and a Volt is the battery. If the battery adds $4,000, that's $28K, or the average new ICE vehicle.

Here's what GM's CEO says about the Volt's costs: “My job is to get it out there and get it right the first time but then get it cost-effective so that we can do a huge number,” he said. “If I had to go with my first generation, we couldn’t really pencil a business case. Any new technology is expensive, but if you get to the second or third generation you find that the cost goes way down”.

Now, I think GM is over-pricing the Volt in order to capture the premium that early adopters re willing to pay, as well as the new tax credit. It's worth noting that he Volt was originally planned to sell at about $30K. Then, the federal government passed a $7,500 tax credit aimed at the Volt, and the expected price of the Volt rose to...the high 30's. Similarly, Mitsubishi plans to sell the iMiev for the high 40's in Japan (that's where the $50K price comes from - it's an estimate based on the price in yen, and is before Japanese subsidies - the price in the US is likely to be very different), where tax credits will bring the price back down to...about $30K. I see a pattern.

Didn't GM raise their expected cost long before a $7500 tax credit was passed?

The tax credit was in the planning stages well before that. The credit is clearly customized for the Volt (some people in Washington call it the "Volt-credit", or something like that) - GM was certainly involved in the planning.

Perhaps their earlier $30k price was just hopeful dreaming at an early stage of the design?

Well, a PHEV is no harder to cost out than an ICE. Keep in mind that EVs have been around for 100 years (GM sold electric trucks in large quantities from 1912-1918); that there are probably 100 million electric vehicles in use, albeit not highway legal; that electric motors are ubiquitous and extremely well understood; that GM designed and built the EV-1, which supplied much of the technology of the Volt; that GM has enormous experience in large electric drivetrains in electro-diesel trains; and that electric drivetrains are much simpler than ICEs. Heck, whenever a new generation of Prius came out, GM would tear it down and cost it out, down to every component. They know how to cost these things out.

Perhaps as the design got fleshed out their estimated break-even cost rose above $40k?

Yes, and they said the reasons were: unexpected costs to get high energy efficiency components ready for the 1st gen; and inclusion of the cost of two batteries, just to be safe. Even if you believe that GM really thinks they need to plan for 2 batteries, these aren't costs that will exist for later models.

And hasn't that estimated price gone as high as $48?

Well, Lutz once said in an interview that they might have to charge $48K to make a conventional profit. But again, that's for the 1st gen Volt. And, we have to remember that pricing is an artifact of accounting: GM is spending about $1B on R&D for the Volt: if you allocate that to the first 50K of vehicles, that's $20K per vehicle. If you allocate it to the fist 1M vehicles, it's only $1k.

Keep in mind that an EV is much simpler. It will cost substantially less when produced in volume. Look at the Prius: it costs about $24K, and it has two drivetrains.