November 2008 Archives

Lies, Damned Lies, and Plug-in Hybrids

Gas/Electric Hybrid Vehicles

About ten years ago, the Toyota Prius and Honda Insight entered the US car market and have grown to change the way we think about automobiles, the environment, and energy efficiency. Starting slowly at first, sales of the Prius took off and now they are one of the most popular models sold in the US.

These hybrids are simple to understand: they run on gasoline just like every other car on the road, but they have a battery pack and electric motor that makes them more efficient: they get very good gas mileage.

But it's an odd design to meld two drivetrains into a single vehicle. Why is it more efficient to make a gasoline engine push around an electric motor and battery pack, and also make an electric drive push around the gas engine and fuel tank? It's not clear to me that it is that efficient. If you look at the top fuel efficient 2009 vehicles according to the EPA, you'll see that the top vehicles are hybrids, but their advantage is mostly in city driving. The diesel Jetta is just 10% less efficient than the Prius on the highway. In Europe, there are even more efficient diesel vehicles.

Hybrids work well for city driving because they use regenerative braking to capture some of the energy that is normally just dumped into wearing out your brakes when you slow down for a stop light. Even though only a portion of that wasted energy gets stored in the battery pack, it's enough of an improvement to make the double-drivetrain vehicle more efficient.

Hybrids are able to offset some of the weight of the electric drive by using a smaller gas engine. The electric drive can help push the vehicle up a hill, and get some of that charge back on the down slope.

Gasoline engines are only about 25% to 30% efficient. That is, only about 25% of the energy contained in a gallon of gas makes it to the wheels to propel the car. The rest of that energy is wasted as heat and mechanical inefficiency. A good part of that gets wasted in the transmission because a gas engine only produces high power/torque in a narrow band of RPMs, so multiple gears are required for good acceleration at a wide range of speeds.

An electric drivetrain can be over 80% efficient. There's no heat wasted in exhaust and no reciprocating pistons. Also, an electric motor can deliver high torque and power over a very broad RPM range, so there's no need for a transmission and thus no mechanical losses there. That's how adding the weight of a second drivetrain that is just fed with a fraction of the kinetic energy normally wasted by braking can improve the efficiency of a gas engine in city driving.

Plug-in Hybrid Electric Vehicles

If that little bit of saved energy can be used to create a more efficient vehicle, wouldn't it be even better to use some grid electricity to further increase vehicle efficiency? Power plants generate electricity more efficiently and cheaply than using a gas engine to generate electricity indirectly through regenerative braking. So, maybe we should further augment a hybrid's power with grid electricity.

That's a promising idea, and is the basis of plug-in hybrid electric vehicles, or PHEVs. There are actually two PHEVs that are generating a lot of buzz now: the Hymotion Prius upgrade and the Chevy Volt.

Hymotion created an after-market upgrade that turns a standard Toyota Prius into a PHEV by giving it an additional battery pack that can be charged from an ordinary outlet.

The Chevy Volt has an even more innovative design: it has a pure electric drive, only the electric motor is connected directly to the drivetrain. It also has a small, gasoline-powered generator that is only used to recharge the battery pack. Because the gas engine is only used as a generator, it can run at its most efficient power level and avoid the gross inefficiencies associated with a car's engine that has to run a wide variety of RPMs and load levels outside its most efficient power range.

Lying about Efficiency

The PHEV is a surprisingly more complicated solution in part because we have no way to talk about the efficiency of this type of vehicle. We're used to evaluating vehicle efficiency by looking at miles per gallon. That works great with a hybrid, because the only energy input is gasoline, but what about a PHEV? The easy thing is to just quote an MPG number and move on, but that doesn't tell you anything.

Consider a different case. Suppose I invent a new kind of hybrid vehicle: gas and propane. It has two engines, a conventional gasoline engine and a propane engine. Together, they power the vehicle's drivetrain. When I take my new model into the EPA to get its fuel efficiency rating, I fill up both tanks. The EPA drives the vehicle on their standard course and find that the car traveled 200 miles and used two gallons of gas, so it gets an EPA rating of 100 mpg.

But what about the propane? How much propane did the car use up? How much does that propane cost? How does the use of propane and gas change with different driving conditions? We already have city and highway numbers, but maybe this new hybrid is even more complicated. How does the hybrid bit work, does it burn propane until it runs out, then switches to gasoline, or does it burn both equally over the entire range? How are consumers going to evaluate what it will cost them to drive this vehicle on their daily commute. How will environmentally-minded consumers evaluate its overall energy efficiency and carbon footprint?

Obviously, was can't just quote an MPG number for a hybrid vehicle that takes in two different fuel/energy sources. That would be misleading. In fact, unless the MPG number works in all driving scenarios, it would be fraudulent.

The same issue applies with PHEVs. If we just get an MPG number, that tells us nothing useful unless we understand how the trade-off between gas and electricity works under our individual typical driving conditions.

The Volt and the EPA

Consumers will want some sort of fuel efficiency number and consumers understand MPG, so GM talked to the EPA and argued that the EPA should use a testing regimen that will give the Chevy Volt a rating of over 100 MPG. The problem is that if the EPA allows the Volt to use the battery pack without accounting for the extra energy input, it gets over 100 MPG, but only about 48 MPG if they don't allow it to deplete the battery pack. The truth perhaps lies between these two numbers and depends on an individual's driving profile.

It's really important that the conversation doesn't stop with this one deceptive measure of fuel economy. The Chevy Volt can go 40 miles on just electricity. That's great if my daily commute is under 40 miles (and that's true for 78% of personal travel in the US according to a 2003 Department of Transportation study), but if I go over that, is it the same as driving a Prius? Unlike the Prius, the onboard engine isn't powerful enough to power the car, it can only add charge to the battery pack. If you just keep driving, eventually the battery pack will run out, and simply filling up the gas tank doesn't refill the source of power that drives the wheels. So, how far can you go? The answer is going to be complex since the gap between what the car pulls from the battery pack and what the generator puts in depends on the speed you're driving. That's not an issue with the Prius, but it's something potential Volt buyers need to understand.

The same issues apply to any PHEV that uses a small gas engine only as a battery-charging generator.

The Hidden Cost

Not only does MPG not tell us enough about how much gas the car uses, while also skipping over the cost of electricity,* it completely hides the cost of the huge compromise built into a PHEV.

The very best battery technology available today is called lithium-ion. This battery chemistry has the best balance of cost and energy density. For a given weight in batteries, lithium will allow you to store the most charge at a reasonable cost. And the cost isn't cheap, either. Lithium ion batteries are more expensive than lead-acid (like your regular car or boat battery) or the nickel metal hybrid batteries used in hybrids like the Prius.

None of these battery chemistries used in vehicles like to be overcharged or fully discharged. If you've ever left your lights on overnight and not only drained your battery, but also ruined it, you know what I'm talking about. With an electric vehicle, there's a computer that monitors battery charge state and keeps you from damaging the batteries, so you don't have to worry about it, but it does have performance implications that prospective buyers need to know about.

Consider a pure electric vehicle like the Tesla Roadster. It has a large pack of lithium ion batteries, big enough to support an EPA verified range of 244 miles (mixed city and highway). Since most commutes are far less than this, 78% under 40 miles and 92% under 70 miles, this means most driving in the Roadster will only need to use the middle of the charge range: it doesn't need to be fully charged nor fully discharged to handle daily driving. This is the best way to ensure maximum battery life. If a Tesla driver frequently uses the entire maximum range of the battery pack, the lifetime of the battery pack will be shortened. The Roadster is not your best choice as a road trip car. Fortunately, road trips represent a small fraction of travel in the US, so this isn't a problem, just something to think about when you're choosing between the Prius and the Roadster for that big road trip.

But what about a plug-in hybrid, like the Chevy Volt with a 40-mile electric range? Obviously, GM has to keep the battery weight down since the car is already packing two power plants. The Volt is designed and marketed as being pure electric for a 40-mile daily commute. If GM were to put in a battery pack that could just barely manage the forty miles, then drivers would put a full charge cycle on it every day. That would kill a lithium ion battery pack in about two years. Let's assume they want their product to last longer than that.

Since the car is designed to be gas-free for a 40-mile commute, that battery pack has to be capable of much more than just 40 miles while also bearing the burden of pushing around a gas engine, generator, fuel tank and exhaust system. So, GM decides what an appropriate charge capacity margin is, and puts in a battery pack that large.

Let's suppose they only want to use the middle 50% of the charge range, so the battery pack is only charged to 75% and only discharged down to 25% (which is about what Toyota uses in the Prius). Based on that assumption, if you drive a Volt on your 40-mile commute, you're going to use half a discharge cycle every day. You bought a battery pack that is capable of an 80-mile trip if you are willing to compromise battery life for an occasional long trip. In fact, if you could pull out all the extra weight of the gas generator, your battery pack could maybe handle a 100-mile trip. Instead, you only get the 40 miles, while still also hammering the battery pack pretty hard, and dealing with all the maintenance hassle of maintaining the gas engine.

Maybe 50% charge buffer isn't the choice that GM makes. If they pick a smaller charge buffer, the battery pack wears out sooner. If they pick a larger buffer, then they are just wasting more battery pack on a hobbled electric drive that could handle even longer occasional pure electric trips. Not matter how you slice it, trying to drive a daily commute with a small battery pack burdened by extra generator weight wastes the full electric potential of the vehicle.

Driving Pure Electric

Compare that to a pure electric vehicle with a 240-mile range. You can do your 40-mile commute with just one sixth of the battery pack's charge cycle, and you have a car that can go over a hundred miles with less impact on battery life than your daily commute in a Volt. Even a 200-mile trip is possible while leaving 20% of the charge range untouched. That's excellent battery life in a vehicle that never burns any gas and is capable of a good long drive, especially if you can get access to an outlet at your destination.

Right now there aren't many choices when it comes to driving pure electric, but that's changing. Just like when any new technology is introduced, initial models are expensive and produced in low volumes. Even the Model T was viewed as a rich man's toy when it came out. With higher production comes both better availability and lower prices. Although electric vehicles have been around longer than gas-powered vehicles, the production electric vehicle market is in its infancy, but is about to get far more interesting.

Today, you can buy a high-end, pure electric sports car with a top speed of 125 mph and an EPA-certified range of 244 miles: the Tesla Roadster, available in limited quantities for a mere $109,000. If they cost less, you probably still wouldn't be able to get one because demand would far outstrip the production rate of about 1200 per year.

But Telsa isn't in the business of solving a shortage of expensive sports cars. Their mission is to get lots of affordable electric cars on the road, the Roadster is just the start. In 2011, just months after GM is expected to start producing the Volt, Tesla Motors expects to start delivering their $60,000 Model S, a luxury sport sedan with a range of about 240 miles. By 2012 or so they expect to deliver their third model, a $30,000 all-electric economy sedan.

But Tesla Motors isn't the only one in the game. Lots of companies, both big auto makers and daring start-ups are promising electric vehicles in the near future.

Aptera expects to start producing their Typ-1e, an EV with a 120-mile range in late 2008, available initially in California for $27,000. BMW is working on an all-electric Mini-E version of the Mini Cooper, available for lease through a pilot program this year in California, New York and New Jersey. In 2009, Miles Electric Vehicles expects to begin delivery of their highway speed sedan, cleverly called the "Highway Speed Sedan," with a top speed of 80 mph and a range over 100 miles for about $40,000. Daimler has plans to introduce electric versions of both a Smart car and a Mercedes in 2010.

Brother, Can You Spare a Trillion Dollars?

Meanwhile the big Detroit automakers have resisted years of pressure to produce more efficient vehicles, instead betting their profitability on giant gas hogs. Who could imagine that either environmental or national security concerns could sour the American public on huge gas guzzlers? Combine that with the brutally obvious result of global oil production leveling off while demand has continued to grow, literally exponentially. Is it any wonder years of short-sighted profiteering have put the big American automakers on the edge of bankruptcy? All of their lobbying to prevent more stringent domestic fuel economy standards while also locking  more efficient diesel fuel vehicles out of the US market has destroyed their competitiveness overseas, and now the American buyers aren't interested in their bloated product lines either.

Their solution is to have the US Government pour hundreds of billions of dollars into the ailing US auto industry to pay for their past mistakes, while they try to retool to build incrementally more efficient vehicles based on a compromised PHEV design, hiding behind inflated and misleading MPG numbers.

That's not how I want my tax dollars spent.



*The cost to drive a car on electricity is generally really cheap, due to the superior efficiency of an electric drive, even taking into account power plant efficiency and transmission loses. But, the cost does depend on where you live. Also, the emissions associated with the energy used in an electric vehicle vary widely depending on how electricity is generated in your area.

The good news is that we are already motivated to green up our electrical generation and EVs benefit from that without changing the car at all, while their gas-powered peers get dirtier with age. Oh, and gasoline prices can only go up, give or take short term fluctuations: global production is flattening out while worldwide demand is increasing.

About this Archive

This page is an archive of entries from November 2008 listed from newest to oldest.

October 2008 is the previous archive.

January 2009 is the next archive.

Find recent content on the main index or look in the archives to find all content.