Expert: Don’t Dumb Down Window Stickers for Plug-in Cars

By · August 09, 2010

Mike Duoba testing a Tesla

Mike Duoba, chief engineer at Argonne’s Advanced Powertrain Research Facility, looks on as Geoff Amann, senior technician, completes a driving cycle with the all-electric Tesla Roadster at the Lab’s two-wheel dynamometer laboratory.

For more than a half-century, consumers have associated vehicle fuel consumption with a single number: the federal government's Miles Per Gallon rating. When the window sticker reads “50 mpg,” consumers know it’s an amazing fuel-sipper. And when it reads “10 mpg,” you know it’s a gas-guzzler. But in a world where drivers charge cars up with electricity instead of pumping up with petroleum, MPG is meaningless.

Okay, but what number can replace the MPG window sticker for a plug-in hybrid or electric car?

After three years of looking into the question, task force committees organized by the Society of Automotive Engineers are no closer to a new metric to replace MPG. That’s because, according to Mike Duoba, a research engineer at Argonne National Lab who serves as chair for the SAE’s J1711 and J1634 committees, “There is no final answer.”

Duoba speaks about the challenge in philosophical terms. “You just can’t explain to a one-dimensional man what a two-dimensional painting looks like,” Duoba told me.

Magical Thinking

The SAE test protocols will not produce a single number for an individual vehicle. Those three years of hard work, instead, produced a complex test protocol to yield an entire set of data. SAE's J1711 is the recommended industry practice for measuring emissions and fuel economy for hybrids and plug-in hybrids, while SAE J1634 is the test procedure for electric vehicle energy consumption and range.

Each car is tested on five different test cycles, including urban and highway routes of different lengths, and cycles for cold temperature, heavy loads, and using air-conditioning. In addition, results for plug-in hybrids with varying degrees of all-electric range and battery sizes are weighted based on large Department of Transportation surveys of average U.S. driving patterns.

The responsibility for transforming all that data into a meaningful number or set of numbers is falling on the Environmental Protection Agency. With the first plug-in cars—the Chevy Volt and Nissan LEAF—only months away, the EPA is running out of time to figure out what will be printed on the window stickers.

Some observers suggest that the EPA use some kind of electric MPG proxy—a fabricated term called “miles per gallon equivalent” or “MPGe.” Duoba doesn’t like that idea for multiple reasons. After years of working on the challenge, Duoba believes that “trying to boil the information down to one number is only going to mislead people. There’s no magical way to turn electricity into gasoline.”

Getting High

There are at least two other reasons not to dumb down plug-in car efficiency to a single answer. First, the numbers are going to be too high to make any sense. Last year, G.M. advertised that the Chevy Volt would achieve 230 miles per gallon, while Nissan said the all-electric LEAF would get 367 mpg. “Any time you go over 100 miles per gallon, depending on how you’re calculating things, you’re going to really confuse people,” Duoba said. “The difference between 200 miles per gallon and 300 miles per gallon can fit into a thimble. As the numbers get higher, the amount of fuel we’re talking about is tiny.”

Duoba also worries about “blatant gaming or unintentional gaming of the system by carmakers.” He believes that a carmaker could design a car to turn on its engine only once, for a second of two, during a specific test cycle. “You average that out, and it will be 900 miles per gallon. That sounds like a really impressive number because miles-per-gallon is upside down. MPG is an incorrect metric because consumption is what really matters.”

Raising the Bar

While Duoba believes that a direct MPG or MPGe number doesn’t make sense, he acknowledges that consumers understand that a vehicle with a certain MPG has a certain level of impact on an owner’s wallet, the environment, or the nation’s dependence on foreign oil.

For that reason, Duoba believes that the results of the SAE protocols could produce a table to indicate how a plug-in hybrid or electric car would fare in terms of fuel cost, petroleum consumption or CO2—compared to an internal combustion engine.

Wouldn’t that require consumers to think deeper about energy consumption?

Yes, and that’s perfectly fine, according to Duoba. “Don’t dumb it down too much. I’ve heard people say the window sticker has to be understandable by a fifth grader. Well, fifth graders don’t drive,” Duoba told me. “We need to place the bar a little higher, folks. If you’re an early adopter, you can have more than one thought in your head at once.”


· George Parrott (not verified) · 8 years ago

Is it too obvious to consider a posted use cost as "dollars per 100 miles?" This could be reported against a fuel cost for either regular or premium as needed by the car for that distance or with electricity or electricity AND fuel for that distance for ER-EV type propulsion.

For a typical gasoline driven vehicle that cost might range from $6.00 (for a Prius at $3/gallon and 50 mpg) to around $4-5.00 for a Chevy Volt (getting the first 40 on EV at around .10 per kw charging cost) and bottoming out at around $2.00 for a Nissan Leaf at $.10 per kW charging cost....

Though my numbers are certainly NOT accurate, they are in the ballpark for the differential actual "use costs" of the various types and levels of power train. Heck the posted number for a Bentley or Hummer would be around $30.00 and certainly communicate CLEARLY the routine costs of the various driving options.

What am I missing here?

· Jim Mapes (not verified) · 8 years ago

The proper way to express this would be KWh/Mile. Converting MPG into KWh/Mile based on the car's HP rating etc would then give a basis for comparison between a gas powered car and an electric one. In other words, if I took the gas and converted it to electricity, how far would I go? The galon of gas would have a rating of KWh of energy, that energy can move a vehicle.

· · 8 years ago

That sounds about right, and feasible. Cost is a major factor, for sure. We have some numbers from current EV drivers, and will have a lot more from Volt and LEAF owners (from various parts of the country) soon enough. I'm thinking that this is a case where window stickers will be a lot less useful than websites with reported data from drivers.

For some people, petroleum displacement and CO2 will be the bigger factors to understand. From my POV, Mike Duoba hit the nail on the head: Cost, petroleum use and CO2 will form the metric matrix.

· · 8 years ago

From my interview transcript with Mike Duaba:

BB: What’s your sense of a cost per mile number?

That’s exactly the approach you need to take. By giving MPGe, you’re saying you found a way to give these guys equal footing. There’s the approach where you take the direct first law of energy conversion of gasoline to electricity. Well, if I give you a gallon of gasoline, with any modern technology or alien technology, you can’t convert that gallon into this amount of kilowatt hours. So, why would you say that they’re equal?

BB: But you can convert it to a certain distance of propulsion?

There’s a conversion of 34 kilowatt hours per gallon or something. But the fact is that if I give you a gallon of gasoline, you can never convert that into 34 kilowatt hours per gallon. You’d be violating the second law of thermodynamics. You have thermal energy and mechanical energy. One is an advantaged state. If you want to convert from thermal into mechanical, you got to use kornow (?) cycle, and availability, and temperature conversion. It’s a second law thing.

· · 8 years ago

As a consumer, I would agree that more information is generally better than less information. EVs also ought to have EPA range estimates on their window stickers, in my opinion. As for CO2 emissions, that is highly variable depending on local power mix, so maybe it would be better if those numbers were specific to the state in which the vehicle is being sold, California obviously being much better than West Virginia.

· · 8 years ago

@ George Parrot,
The problem with $/mile (or $/100 miles) is that it assumes an energy cost. Electricity prices vary over a factor of 5 between highest and lowest, sometimes even on the same bill depending on time of use or usage tier.
I'm disappointed with Mike Duaba's statement: "you can’t convert that gallon into this amount of kilowatt hours". Why can't he find out how many kilowatt hours one could convert a gallon of gas (or its raw crude feedstock) into, using an efficient large-scale generator. That could give some estimate. Note that that would probably be close to 70% of the theoretical first law of energy conversion of gasoline to electricity for combined-cycle (co-generation) power plants.
We all know that EVs will be much better than gasoline, we just need some way to roughly quantify it for people.
I don't think that 75% of GM's 250 mpg estimate or 175 mpg is going to hurt their sales any when comparing a Volt with a 30 mpg gas guzzler and it isn't too far off of reality.
Actual kWhr/mile or the equivalent should definitely be stated in order to compare EVs with each other or for estimating electric bill impact or solar installation requirements.

· Kei Jidosha (not verified) · 8 years ago

How about measuring reality?

For EVs, City and Highway numbers for miles/kWh, along with City and Highway range. “Highway” should be measured using the EPA US06 test, “City” should be from the EPA LA04 cycle.

So for battery only mode (generator locked out on Hybrids);
-Mi/kWh Highway
-Mi/kWh City
-Range Highway
-Range City
Multiple (5) recharge cycles should give reasonable accuracy.

For Hybrids, MPG measured with battery at charge cut-off minimum state-of-charge at start and finish of the test;
-Mi/Gal Highway
-Mi/Gal City

I don’t think this would take 3 years, and the results would mean something.

· Freddy (not verified) · 8 years ago

We won't have to worry about the EV van catching fire in an accident like with ICE engines.

If they can get the post office and municipal government to use the Transit Connect van that would give some revenues to Ford -Azur to keep building these vans and it would be great advertising for the Transit Connect vans.

They need to put a solar panel on the roof to trickle charge the battery

And they need to paint the vans in better colors to attract more customers.

· Freddy (not verified) · 8 years ago

I forgot to say that they need to have 2 modes:

1 - If you release the accelerator pedal, the van is slowed down by the drag of the electric motor, thus saving on the brake pads and regenerating the battery by reverse charging.

2 - coast mode: when you want to go down and up the next hill with out pressing on the accelerator till you're half way up the hill.

If you want women to buy these vans to drive around their brats,
you need to sell them in gold, green, tan. White only won't cut it.

· Freddy (not verified) · 8 years ago

One more:

I watched the DVD : "who killed the electric car" last night.

You might be able to find it at your local library like I did.

Excellent explanations about the EV1 conspiracy.

Why doesn't GM not just recycle this EV1 car ???

· Freddy (not verified) · 8 years ago

Most consumers don't look at these numbers.

It's the style, the color palette and the feeling when they drive it that matters to them.

Consumers have a monkey see monkey do way of buying.
If they see others driving an electric car and liking it; then they will want one to be better than their neighbors.

They don't have the brains to analyse all these numbers.
Except... the price tag. $$$

· · 8 years ago

Just a comment about "converting" gasoline to electricity. Yes, there are conversion factors using 1st Law Thermo conversion of gasoline to kWh, depending upon the actual gasoline batch it hovers around 34 kWh / gallon. Here is the rub, if I gave you 1 gallon of gasoline, you would NEVER be able to give me back 34 kWh of energy - it's lost in any real conversion process. You can't with any exotic new technology, with any technology 100 years in the future, or even if you had access to alien technology, you will not convert it back into 34 kWh of energy.

What this means is that we can't mix together incomparable measurements. We need to focus on common metrics that can be brought into focus for one figure of merit.

Some figures of merit for electrified vehicles:
- Reduce petroleum usage (essentially fuel-only consumption rates)
- Reduced consumption related costs (fuel and electricity)
- Reduced CO2 emissions (full fuel cycle analysis tells the whole story)

And all these metrics must be in context of what percentage of miles are driven in charge-depleting mode compared to charge sustaining mode. Either the consumer figures this by their own estimation, or if thinking of the entire fleet, then one would employ "Utility Factors" (SAE J2841 addresses this).

- Mike Duoba, ANL

· George Parrott (not verified) · 8 years ago

The current "mileage sticker information" also includes, as I recall a "cost per year" for that car figured on a specific noted cost per gallon.

That same kind of "working assumption" could easily be done for the "Dollars per 100 miles" estimate by simply using some "typical" (if not absolutely accurate) cost per kW in the same way as a typical cost per gallon of regular or premium fuel would be set and extrapolated. One simply notes on the info sheet what those "working assumptions" are per kW and per gallon.

I would further remind all of those anticipated future EV drivers that many utilities have VERY special and reduced EV charging rates for late night electricity draw. For P G & E that EV charge rate is around $ .045 per kW where the lowest regular Tier 1 utility rate is around $ .09 per kW. I believe in order to get this special EV rate forms have to be filled out and likely an additional meter installed....But the savings seem considerable ....

· · 8 years ago

Ok, I see Brad already added my opinion on Gasoline to kWh conversion.

Yes, one can show how much "energy" came from a power plant to charge a vehicle. However, one critical objective that PHEVs (and BEVs) is achieving is relying on domestic fuel. Power plants use domestically sourced fuel. Perhaps I should clarify the first figure of merit to be Reduced Imported Petroleum Usage.

Electrification is not necessarily an energy efficiency technology, its a petroleum displacement technology. This is why a heavy, powerful Fisker, for example, may not be more efficient than a conventional hybrid, but the huge gains are realized from avoiding gasoline usage.

- MD

· Anonymous (not verified) · 8 years ago

The answer rest on the Volt. If you want to see how far you get on a gallon of gas in an electric car, use the Volt. Drain the battery, place a gallon in the tank and see how far you get. The gasoline is converted to electricity by the onboard generator and propels the car. Test at city and highway speeds. There you will have miles per gallon in an electric car. Then calculate how many kilowatts are used to cover the same distance in all electric mode. You will then have two values on sticker. MPG in gas mode and MPGe (miles per gallon in electric fuel source mode). Then could add a third value, cost to operate on electric vs. gas.

· · 8 years ago

I've mentioned this before when this topic came up.

What's a kilowatt hour? I know, I can be explained and I understand what it is. But in general if you ask the average person about their electric bill and ask how many kilowatt hours they used, they don't know. They will tell you however how much it cost.
Now if you ask a person that has a home heated with oil, they'll tell you without a second thought how many gallons they put in last month.
So what's the problem with watts?
It's that we're so used to measuring our car's energy capacity and efficiency with a liquid volume.
Now we're being to told to measure it with a unit of electricity and time.

A new standard needs to be put in place that people can relate to. Say...a Charge Unit (CU)
The details of what that CU is can be worked out, but it has to be something people see or use every day.

Once that CU is established, people can say for example, "I have a car 10CU battery that gets 50MPCU

· MichaelWoon (not verified) · 8 years ago

A lot of smart, reasonable people on this page! This is great.

Thanks for the interview from Mike Duoba, and thank you for the comments, Mike.

Mike, you state that electrification reduces CO2 numbers, when considering fuel cycle analysis. Is there a specific paper or report that explains this, and is a fuel cycle analysis included in the GREET model? Do you have to make any assumptions/projections on the U.S. grid composition? (i.e. more nuclear, less coal)

I'm pretty sure the 2006 U.S. average grid composition would produce more CO2 per mile than a pure ICE, on *average.* I'm not critical of EVs, I'm saying if people start plugging in their cars to ease their conscience, they better start thinking beyond the plug. When I can afford my first EV, I'll buy the E-Tracer (X-Tracer if you were at XPRIZE), but the second thing I buy is a bank of batteries and a hydro-generator and/or wind turbine. I'll be supremely happy to get my transportation demands in-line with my renewable energy generation. :)

-Michael Woon, UM

· MichaelWoon (not verified) · 8 years ago

Also, the 38.3 kWh AC per gallon of gas is bogus, sort of. To briefly clarify what Mike has been saying, I'd propose what thermodynamic engineers have done for a long time, and that's label the energy.

For example, 34 kWh_therm = 1 gallon of gas (also specify uncondensed, LHV).

Or, today I used 5 kWh_elec.

Most cars need 0.3 kWh_mech per mile for propulsion and cruising.

That way you know if you have 5 apples or 5 oranges, not just 5. And you can't go from one category to another without losing something. For example, the law of thermodynamics states, simply, 1 kWh_therm ~= 1 kWh_mech (not equal).

-Michael Woon, UM

· Daniel (not verified) · 7 years ago

You may have a lot of faith in the buying power of the consumer, but in all honesty, simpler sells. You won't find marketers jumping the gun at complex KWpH calculations and such...
And they aren't needed. You don't have complex gasoline ratios on a sticker for a car, either- even though the composition of the gas and the way it's driven can affect it. You have MPG ratings.
So, you want a simple window sticker replacement that a consumer can understand for an EV?
Miles Per Charge.
Yeah, factors can adversely affect the number- but so can A/C use, driving patterns and gas ethanol content on a regular car, too.
But when a consumer buys an EV, that's what they're interested in- they want to know, without compex calculations and kw anything, just how many miles they can expect to get out of a "fill up" on a car, just like they can expect a certain number of miles to the gallon on a regular one.
Problem solved.

· · 7 years ago

Hey Daniel,
I agree with you about simplicity. Maybe there's a middle ground. Provide a high-level stat or two, plus maybe a grade (as propsed):

And then give the details for those who want to more fully understand.

· Car Window Stickers (not verified) · 5 years ago

Based on the concept you wish to spread through your car decals and your price range you need to choose upon the budget of car decals at the same time

New to EVs? Start here

  1. Seven Things To Know About Buying a Plug-In Car
    A few simple tips before you visit the dealership.
  2. Incentives for Plug-in Hybrids and Electric Cars
    Take advantage of credits and rebates to reduce EV costs.
  3. Buying Your First Home EV Charger
    You'll want a home charger. Here's how to buy the right one.