New EPA Numbers Promote Confusion About EV Efficiency

By · March 12, 2012

Coda Sedan EPA Label

According to the EPA, the Coda Sedan has an 88-mile driving range, but an MPGe rating of only 73.

The U.S. Environmental Protection Agency released efficiency numbers for the Coda Sedan, assigning a rating of 73 MPGe. Okay, that's disappointing compared to the LEAF's 99 MPGe, the Ford Focus Electric's 105 MPGe and the Mitsubishi i at 112 MPGe. On the other hand, because the Coda has the biggest battery of the bunch, it beats all other EVs with an EPA "miles on a charge" rating of 88 miles.

These numbers seem designed to instill confusion.

In August 2010, before any of these ratings came out, I spoke with Mike Duoba, a research engineer at Argonne National Lab who serves as chair for the Society of Automotive Engineers' committees related to EV efficiency. He warned not to dumb down the numbers. Duoba said that MPGe (or miles per gallon equivalent) numbers would be too high to make any sense. “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.” He was talking about GM's campaign at the time to promote a 230-mpg rating for the Chevy Volt, to which Nissan replied that the LEAF would get 367-mpg if measured in the same way.

While the numbers have come down to a more earthly reality, according to the EPA, the difference in real-world cost or driving experience between the electric equivalent of 75 miles per gallon of gasoline versus 100 or more remains difficult to grasp.

Coda comparison

Compared to the Nissan LEAF and Mitsubishi i, the electric Coda Sedan goes further per charge, but isn't as efficient.

With EPA certification now complete, the Coda Sedan is ready to launch. Initially, Coda will only sell one version of its electric sedan priced at $37,250. Coda is planning to hand over keys to the first electric Sedans in the next week.

As an unknown brand with an unproven car with a dull design, Coda faces more than its share of challenges in the emerging electric car marketplace. Unfortunately, the low and confusing numbers from the EPA don't make life any easier for Coda, or for that matter, the entire growing field of EVs.


· Anonymous (not verified) · 6 years ago

Brad, where did you get the chart? I'd like to personalize it with ActiveE informaiton.

· · 6 years ago

Chart can be found here: Search under 2011 for ActiveE

· Steven (not verified) · 6 years ago

I assume there is some equation that permits conversion of the energy equivalent of kilowatts to a gallon of gas. So the m.p.g. equivalent is simply a measure of how far you can go on that ‘gallon of gas’ (roughly 90% of the LEAF’s battery capacity)?

It seems like it would be fairer to use ‘well-to-wheel’ numbers that included the energy required to refine the gas as well. But how would that stack up against the energy required to produce and distribute electricity, say from coal?

· Stephen Taylor (not verified) · 6 years ago

So these Coda range numbers are for the smaller battery pack? My understanding is that the larger battery pack will not be available until the Fall.

· · 6 years ago

For EVs they really should emphasize miles/kWh, which is the direct analog of miles/gallon. Instead, they give (in small print) "kWh per 100 miles" which is rather unintuitive; smaller = better, but most people see bigger = better.

Put mi/kWh in big print in the upper left corner, then they can have MPGe as the smaller number.

@Steven "...some equation..."

Bit of a sticking point, since how much energy you can consider a gallon of gasoline to contain depends on what you plan to use it for and what temperature it's stored at!

For internal combustion engines, we generally use the lower heating value of the fuel. This is what the EPA uses for comparison, that gives 1 pound of gasoline = 18,790 BTU (1) = 5.507 kWh.

The density (pounds per gallon) will change quite a bit with temperature, but is about 6 pounds per US Gallon, so 112,740 BTU/Gallon or 33.042 kWh/Gallon. 113,000 BTU/Gal is a common number so bump that up to 33.12 kWh

So there you go: 1 US Gallon gasoline = ~33.12 kWh electricity.

For well-to-wheels, the US DOE determined that manufacturing and delivering gasoline is 83% (2) efficient. Adjust your values accordingly!



· · 6 years ago

The numbers look pretty straight forward. The font size emphasizes the important items, like MPGe's, money save, range. Really, if somebody's progressive enough and wealthy enough to buy this car, they can probably understand these numbers.

For the rest of us, the word SAVE is in bold next to eight thousand plus dollars. If that's not a motivator then, it's probably a lost cause customer anyway.

Plus it implies that the gas car it's sitting next to has $0 next to the word save. :)

· · 6 years ago

Smidge204: I agree, this is really about miles per kilowatt hour. On the other hand, I think if you took a poll of the average person who knows what a Kilowatt hour is (in 2012), the numbers would be depressing. I think people need time to switch from gallons to watts.

· · 6 years ago

I agree with tterbo. Most people don't know what a kWh is, even as they blindly pay their utility bill each and every month. Most people (even many of my fellow engineers) don't know how much they use in their home, or how much a kWh costs. Everybody knows what a gallon of gas costs.

· · 6 years ago

@ Steven

The EPA uses 33.7 kWh per gallon of gas as the conversion factor. Unfortunately, if you call your electric company to see how much the energy in a gallon of gas will cost you, they won't be able to tell you (unless you express it in kWh). I'd prefer that the big number were in miles per kilowatt hour, because you can call up (or look on your bill) and get the cost of a kilowatt hour.

The GREET tables from Argonne Labs give the well-to-pump (or plug) efficiencies of standard and alternative transportation fuels. In the tables I have, gasoline is refined and transported with 82% efficiency, and electricity is produced and distributed at 38% efficiency (from the grid mix of sources). So an average car, at 22 mpg, could get an adjusted well-to-wheels rating of 22 x .82 = 18.04. A LEAF would get an adjusted rating of 99 x .38 = 37.62. On this basis the leaf looks very good, and will look even better as time goes on (and the grid becomes more renewable). LEAFs charged from solar and wind are of course infinitely better than a car powered by gasoline (or diesel), in both the resource consumption and emissions perspectives.

The EPA site provides a GHG calculator which will, I assume, be available for any electric car (but now only covers the LEAF and Volt). You plug in your zip code, and it calculates your GHG emissions.

· Smithjim1961 (not verified) · 6 years ago

I posted the following comment in another thread but this article is the perfect place to repeat it. I had a conversation with a friend about EVs. I told him that the Mitsubishi i gets 112 MPG equivalent. I told him that MPGe derives from the fact that gasoline has the chemical equivalent of 33.7 kiloWatt-hours of electrical energy. Then I said, "You're probably wondering how much electricity costs compared to gasoline" I said the 33.7 kW-h number can be used to compare the cost of electricity to gasoline and that electricity in the St. Louis area is eight cents per kiloWatt-hour which comes out to $2.80 per "gallon" of electricity. He was glad that I explained it to him in those terms and now he "gets it".

· · 6 years ago

In the greater scheme of things, we need both an energy per mile number and a range number in order to compare one car and another. And the reality is that people need to compare ICE cars to plug-in hybrids to EVs in some meaningful way. I think the only way to do that for the public at large is to list both the MPGe and the kWhr/mile even though they are reduntant for the energy per mile part. The MPGe allows a plug-in hybrid or EV to be compared to an ICE in at least a rough energy usage sense. And it allows people new to plug-ins to readily understand how much less energy an EV would use. Redunantly listing the kWhr/mile allows the prospective EV buyer to calculate energy costs in their area by looking at their utility bill and prepares us for the future when all cars are rated that way.

Listing the Coda MPGe and range does go a long way to describing what the Coda is compared to the LEAF, Volt or i. The Coda is not as energy efficient as these other vehicles but has a big enough battery pack to still get a longer battery range anyway.

That is actually useful information for someone to have when choosing which vehicle to buy. Anyone interested in minimizing energy use would be better off with one of the other vehicles. On the other hand, anyone just wanting to not use gasoline and needs the additional range might consider a Coda. Different people have different needs and desires and so having both pieces of information is useful.

The EPA sticker may not be the best that it could be, but at least it tells people what they should know to make a rough comparison. I personally dislike the kWhrs/100 miles and would prefer just kWhrs/mile, but they are just two different units for the same thing and thus work as well for comparison sake. It just makes it a little more confusing for someone unfamiliar to look at their bill and figure out their cost per mile.

· · 6 years ago

The problem with using kWh/mile is that it is a fraction, perhaps a number like 0.3. It may sound silly, but that is harder for many people to grasp than a number like 30 kWh/100 miles or a number like 3.3 miles/kWh.

The scientist in me hates the MPGe number because it isn't really meaningful, but it does have the advantage of putting ICE cars, PHEVs, and EVs on the same scale for comparison, however crude.

· · 6 years ago

The MPGe number is only a temporary, transitional number. It is useful to people during the transition. Later, it will naturally disappear.

· The Champ (not verified) · 6 years ago

Making the rating a MPG number to me is not really the right thing. Perhaps an "Average Miles per Battery Charge". Then compare it to an average car that is in the same size class. Sort of like they do on a new refrigerator. Then based on a rate per KWH, give the estimated cost per year based on 12,000 miles or something. Then compare it to the same cost per year for that average car in its class, the best other car, and the worst.

· · 6 years ago

"The MPGe number is only a temporary, transitional number. It is useful to people during the transition. Later, it will naturally disappear."

Kinda like how the the term horsepower went away. :-)

· Mike Duoba (not verified) · 6 years ago

The problem is, the MPGe for an EREV or BEV is different than the MPGe of a Prius-style PHEV which is still different than the MPG of a conventional vehicle. So why do we have MPGe? Because the EPA law (written a long time ago) says that all alternative fuels must be reported in MPGe. Presumably so that they can be compared. But this is the worst thing to do for plug-in vehicles because we cant compare them if they are all calculated differently and mean different things.

I propose reporting only in metrics of concen. For example:
- PHEV/EREVs should be given in MPGa (actual fuel consumption). The metric of concern here is petroleum consumption.
- Also, cost per mile can be useful to compare all vehicle types in all modes.
- Dont use "per 100 miles" for PHEVs, that is implying that it occured over a 100 mile trip, but this is not the case.

As far as the Coda results. Compared to other BEV results I have seen, they get a high score for range, but a lower score for efficiency. But with cheap electricity not based on imported petroleum, the value proposition for the consumer may be on the side of more range = most utility.

· Anonymous · 6 years ago

@indyflick - Usually LEDs list their brightness in lumins, but I always find it funny when they occasionally use candlepower. It brings up images of someone in a castle comparing the brightness of an LED with a candle. Some of these old units just won't die :)

· · 6 years ago

You know what you could do in the meantime, come up with a conversion table for the EPA data and make an EV community standardized kW/m sticker for these cars. Granted the car companies might complain if they didn't like the numbers.

Also, one nice feature of the EPA sticker. If you're figuring on keeping the car for 10 years, the x2 math on the cost savings pretty simple. I like the fact that you can walk up to a car like this and quickly see that it will save $16k+ over 10 years of usage. The savings easily covers any battery replacement cost as well as part of the next car.

· · 6 years ago

@indyflick - that message about the LEDs was from me. I guess I wasn't logged on.

· · 6 years ago

We shouldn't let our desire to be overly precise get in the way of the general consumer figuring out what to make of these new fangled battery powered cars. They need something that brings home to them in a few seconds what the real impact is of what they are looking at. Using MPGe for an EV results in numbers like 78 MPGe to 112 MPGe. This allows a consumer to immediately see that in addition to an EV not using any imported oil, the EV is also using a lot less energy to travel a mile than a traditional ICE car is.

To try to get overly precise about it in those few seconds for a previously uneducated consumer would be an impossible task. There are so many issues in comparing the impact of one thing to another. It depends on what criteria is important to a person and it depends on the detailed background of where and how the car will be used, like the % of coal for grid generation and a number of other things.

The important thing is that the consumer unfamiliar with EVs can very quickly see that there is a big difference here between the EV, the plug-in hybrid and the traditional ICE and that if they try something new the impact they will make will be significant. And that they have the tools to compare different EVs against each other in different ways so they can choose the best one for their needs and priorities.

We have to keep in mind that different people have different agendas that bring them to consider plug-ins. For some it is carbon foot print, for others dependance on foreign oil. For me I don't even care about the impact today because today there are not that many EVs out there. I care about getting the world on a path to a fully sustainable future. EVs sold today enable us to get on that path.

The numbers now on the EPA sticker, no matter that they aren't perfect, allow people to at least have a guild to make a decision based on whatever criteria drives them. Engineers who want to get into greater depth don't need the sticker to do so.

· · 6 years ago

The biggest downside I can see is the bigger the battery pack the lower the MPG rating. You could drive a Mitsubishi I get the best MPG rating on one of your cars and be doing a heap of gas miles on your other car. While if you bought the model S with a huge battery pack it could be the only car you ever need and you could not do any gas miles ever again.

· · 6 years ago

Why do you think bigger battery packs will get lower MPG ratings? I'm guessing that the Model S will get better MPGe than any other EV.

· · 6 years ago

@ex-EV1 driver

My reasoning was because of the weight ratio. If the percentage of the weight of the pack compared to the weight of the car goes up then the efficiency should go down. (ie MPG) For instance in the Mitsubushi because it is quite small if it had a battery pack that was 20% the weight of the car compared to say the CODA whose battery pack was 30% the weight of the car then the Mitsu would be more efficient. (My figures just for comparison)

Maybe I am having a blond moment as I looked at figures for the Model S. The two bigger pack options have 100MPG figures though I am not sure if they are Teslas own figures. One car having a 65kW-h pack the other an 85kW-h pack.

I must admit I am having trouble with this as I would have thought the lighter pack would give better MPG if everything else is the same.

· · 6 years ago

A heavier car will have worse MPGe if that is the only thing that is changed. But there are so many other things that come into play. Like the cross-sectional area and Cd which together determines the aero drag. Or the efficiency of the motor and other electronics. And the often overlooked parasitic losses (like coolant pumps, if they exist).

In the case of the Model S and its three battery size versions, there are some differences in the electronics with the smaller size battery and I don't think Tesla has yet said what they are. Additionally, the lower size battery version may very well use an older battery chemistry. I don't think Tesla has yet announced whether or not that is the case. Last I heard they were still debating that. Different batteries can actually have different internal impedience which can also hit the overall efficiency more for some batteries than others.

Some or all of these things are things that Coda clearly didn't do that great on. And even Nissan is working on improving where they are with these sorts of things so that they can get better range without adding more battery cost or weight. Mitsubishi just recently announced that they, like Tesla, have packaged their inverter into a cylinder the same diameter of their motor, looking like an extension of the motor, saving space in their future EVs. And they have said that this new electronic set will be more efficient than previous versions, increasing range. I am not sure when Mitsubishi's new inverter will show up in their vehicles.

Often people overlook all of these other losses because they see efficiency numbers for these things that the manufacturers release that are very impressive. But you have to keep in mind that most of the component level efficiency numbers that people look at are the peak efficiency numbers. In actual use in real duty cycles these components can lead to much bigger average energy loss.

· · 6 years ago

@ Smithjim1961

Of course, your conversation would have been shorter and perhaps easier to remember if you said "Cars like the Leaf and Mitsubishi get 3 or 4 miles per kilowatt hour. That's about 2 to 3 cents per mile in St Louis, where electricity costs 8 cents per kilowatt hour."

After he says "Wow, that's cheap!" you can say, "Yeah, a typical car gets about 20 miles on $4.00 of gas, for about 20 cents per mile."

· · 6 years ago

@decker and alt-e,
Weight on an ICE gets you twice. It hurts you most because of the extra energy required to accelerate. It also hurts you because of the added tire drag. An EV with perfect regenerative braking would require no added acceleration energy because regenerative braking would recoup the acceleration losses while decelerating - proportionately to vehicle mass. Obviously regenerative braking isn't perfect, but it definitely offsets vehicle mass a bit. You still get more tire drag with more mass.
The better drag coefficient of the Model S and likely other systems could conceivable offset this if it is a lot better.

· · 6 years ago

@ ex-EV1driver and alt-e

Thanks for your input I had no idea it was so damn complicated. Would I be right in thinking that the drag coefficient is most important followed by the motor efficiency & battery chemistry. Would these be the 3 most important factors determining the best MPGe (for want of a better word)

Also how much does (battery) weight affect efficiency.

· · 6 years ago

@Deckard - As ex-EV1driver said, weight increases the energy required to accelerate and increases various friction losses. These friction losses and acceleration are the dominiant losses at low speeds or in city driving where there is a lot of start and stop. The frictional forces are mostly proportional to the speed that the car is traveling, which means that the energy lost goes up with the square of the speed.

The aerodynamic drag force is proportional to the square of the speed which means that the energy loss is proportional to the cube of the speed. Air drag is not affected by the weigth, it is the cross-sectional area times the Cd. The air drag energy loss is very low at low speed compared to the various friction losses, but since it increases faster with greater speed it overcomes the mechanical friction at around 30 mpg, depending on the design of the car. By the time you get to highway speeds air drag is what it is all about.

But the above are just the losses that the car experiences no matter what the drivetrain is. Internal to the drive train there is the motor inefficiency, any electronic power processing inefficiencies (such as the inverter and/or motor controller) and other parasitic losses (such as a coolant pump or the radio or the a/c, etc.). There is also the internal impedence of the battery.

Regen breaking has to encounter the powertrain losses twice. Once when the regen is capturing the kinetic energy to slow down the car and a second time to accelerate the car. This is why coasting is always better then regen when there is a choice.

So you can see that weight is more important in city driving and air drag is more important in highway driving. And the efficiency of the powertrain and the rest of the car's systems are important all the time. Especially in regen.

· · 6 years ago

I meant "30 mph" instead of "30 mpg" above.

· · 6 years ago

@ alt-e

Thanks for that now I understand

· · 6 years ago

@alt-e, "This is why coasting is always better then regen when there is a choice."

This is not true in some circumstances. As Darrel will remember, we had this discussion awhile back: coasting at or near terminal velocity is less efficient than using regen. This is because at terminal velocity deceleration due to drag = acceleration due gravity, meaning that all potential energy is being lost to drag. Slowing down and recapturing some of that potential energy is more efficient.

The example I used was this: assume descending a long hill that allows the car to accelerate to terminal velocity for an extended period. When the car gets to the bottom it will coast a certain distance before coming to a stop and that distance will be proportional to the kinetic energy of motion at the bottom of the hill. Now suppose that the car uses regen to reduce speed most of the way down the hill then, near the bottom, coasts up to terminal velocity. The car will travel the same distance but, in this example, will have recovered some of the potential energy via regen rather than have lost it all through drag.

Another example that I encounter every day, given where I live, is when descending hills with sharp turns. Coasting isn't safe and using regen to keep speeds at a safe level is more efficient (the alternative being friction braking as the turn approaches because the coasting speed is too high to make the turn).

· · 6 years ago

While it is possible that the Model S will be more efficient than the shorter range EVs currently available I am skeptical. Even if the Model S has a somewhat improved Cd the sheer size of the car will mean a larger frontal area. Unless the Cd is enough lower than the current car models to overcome the larger frontal area the Model S figures to be less efficient from a drag perspective. That leaves making up the difference from improvements in charging and drive train efficiencies. Could happen, I suppose, but it remains to be seen.

· · 6 years ago

@dgpcolorado - Naturally it is often the case that regen braking, or even friction braking, is necessary to reduce the speed of the car for safety depending on the driving conditions.

It would take a pretty extreme downhill for regen to be a more efficient choice than coasting, but where you are that extreme case may be common :)

· · 6 years ago

@ alt-e:
A couple minor tweeks are required in your very good explanation.

"The frictional forces are mostly proportional to the speed that the car is traveling, which means that the energy lost goes up with the square of the speed."

Frictional forces are, for the most part, nearly independent of speed. A .006 Crr tire on a 3000 lb car creates 18 lb of drag at both 30 and 60 mph. (The curve is not perfectly flat, but close enough to be considered flat for speeds up through ordinary highway speeds.) That means that at 60, the power required to overcome tire drag is twice that required at 30. The energy consumed, however, to travel a given distance is the same. For example, if 3kW is required for 30 mph, and 6 kw is required for 60, then 60 miles traveled requires 6kWh in either case (travel time being 2 hours in the first case).

(This principal is one influence on the fact that even electrical cars get roughly similar city and highway MPGe numbers -- even though the power required for high speeds, due to aero resistance, goes up so dramatically.)

"The aerodynamic drag force is proportional to the square of the speed which means that the energy loss is proportional to the cube of the speed."

Similar thinking applies here. The force varies with the square of speed meaning that the power requirement is proportional to the cube of speed. Therefore, the energy consumed (to overcome aero drag) is proportional to the square of the speed. So, for example, if a car requires 3 kW to overcome aero drag at 30 mph, it will require 24 kW to overcome drag at 60 (This would be a big boxy car.) 60 miles would require 6 kWh for the first case (2 hr x 3 kW) and 24kWh in the second. (Four times the energy, not 8 times the energy.)

"The Car Equation" ( defines the tractive force required to move a car for any instantaneous condition. This force times the distance moved = energy.

· · 6 years ago

@ · Ken Fry, What a terrific explanation of those concepts! I always try to be careful to use "power" and "energy" correctly because I am aware of the difference. But I never got around to backing energy out of the drag equation to find that it is the square of velocity. Although it makes sense in retrospect, since if drag increases as the square of velocity (roughly) then the energy needed overcome aerodynamic friction ought to also increase as the square of velocity.

When trying to explain why gas (or electricity) mileage improves at high elevation, I just tell people that there are "fewer air molecules to push out of the way". That seems to make more sense to most people than the mysterious term "drag".

· Larry, Richmond VA (not verified) · 6 years ago

440 watt-hr per mile is ridiculously high for city driving. How can it possibly be that inefficient? I did better than than in my 1980's vintage lead acid EV.

· Anonymouse Volt (not verified) · 6 years ago

Yeah, my Volt does 300-330 watt-hours per mile. 440 is astonishingly close to SUV territory.

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.