BMW Builds The Most Efficient “Purpose Built” EV Yet – The i3


BMW i3 Efficiency

2014 BMW i3 – Efficient For Sure…But How Efficient?

One of the conversations that is rarely if ever included in the EV forum discourse is efficiency. There is as much conversation as you could ever want about range, conveniences, charging protocols, acceleration, regen, interior sound levels, battery life, interior capacity and more but very few words are exchanged about how efficiently EVs propel themselves and their occupants from destination to destination.

Even, one might say especially, in the realm of alternative energy vehicles where reducing the carbon footprint is central to the concept, efficiency remains the true premium since in their presently generated form, all alternative fuels have a carbon cost. Possibly someday in the future we will be swimming in gloriously abundant photovoltaic electricity and the story will be different but we are nowhere near that yet.

BMW i3 Engineered For Highest Efficiency

BMW i3 Engineered For Highest Efficiency

True, EV drive trains are inherently more efficient than their ICE counterparts but they are currently powered by batteries and energy storage density in batteries is much lower than gasoline as we know. For a given vehicle package with current technology, EVs are heavier than ICEs due largely to battery mass. As we also know, mass is the enemy of motion, or more accurately change in velocity – acceleration, braking, turning, suspension response and handling are all negatively impacted by mass.

The lower battery energy density makes the balance between mass and performance more delicate for EVs than for ICEs – a few more gallons of gas won’t penalize you much in mass but it will provide substantially more energy. Obviously energy can be increased by a larger battery as well but batteries are heavy and a bigger battery necessitates a stronger structure (read heavier) and the return is not linear. The heavier structure also requires stronger (heavier) suspension components which then require a bigger battery which mandates a stronger structure . . .

The lower battery energy density makes the balance between mass and performance more delicate for EVs than for ICEs – a few more gallons of gas won’t penalize you much in mass but it will provide substantially more energy. Obviously energy can be increased by a larger battery as well but batteries are heavy and a bigger battery necessitates a stronger structure (read heavier) and the return is not linear. The heavier structure also requires stronger (heavier) suspension components which then require a bigger battery which mandates a stronger structure . . .

BMW i3 drivetrain production

BMW i3 drivetrain production

So I thought it would be good to look at EV choices relative to efficiency to see where the market is. I compiled a list of the EVs that have been tested by the EPA with their five cycle test, hoping to result in a data set that is more objective then by using OEM numbers or magazine tests. I recorded the efficiency in kWh per 100 miles for each vehicle from the EPA list (I am only analyzing BEVs so I just don’t want to look at MPGe, too much ICE in it for me). Since I am interested in the BMW i3, I appended several projected performance configurations to the list as follows:

  • BMW i3A – BEV, 90 mi. range, 18.8 kWh batt. (optimistic projection)
  • BMW i3B – BEV, 80 mi. range, 18.8 kWh batt. (more realistic projection)
  • BMW i3C – REx, 75 mi. range, 18.8 kWh batt. (I dunno, just a guess)
  • BMW i3D – BEV, 80 mi. range, 22.0 kWh batt. (the full batt. capacity)

The projected/unconfirmed numbers are in blue text in the tables.

I then calculated Miles per kWh, the metric I like best for BEVs, in another column. I added a few other data points that may be of interest to this discussion and populated them from the OEM websites, 0-60 times came from except when it was available on the OEM website. Sorting on Miles per kWh generated the table below:

Sorting on Miles per kWh generated the table below:

In this analysis, the BMW i3 is the hands down champ in all four  projected performance configurations with i3A more than twice as efficient as the RAV4 and nearly twice as efficient as the Teslas. Assuming the i3 actually tests at one of these levels, the Bavarians have achieved quite a remarkable result – even the worst case projection, the i3D, is at the top of the “do more with less” rankings. But does it sacrifice in other important areas?

As important as it is, efficiency at the expense of other critical performance criteria does not a great product make so I checked out a few other rankings starting with 0-60 times:

The Teslas dominate this measure as we all know. The four i3s hold their own though, high on the rankings, right after the Teslas and the RAV4. The RAV4 is a surprise but it does have that big battery.

Next was Curb Weight:

Pretty Hard To Find Anything Lighter Than A smart ED

Pretty Hard To Find Anything Lighter Than A smart ED

The i3s are not the lightest of the group which may not be surprising considering the diminutive nature of several of the offerings.

As small as the small ones are, including a two-seater, as a group they operate on the lower end of efficiency spectrum emphasizing that mass is not the only factor in doing more with less. The Teslas are the heaviest by a substantial margin – the SP85 is literally a full ton heftier than the i3 BEVs. That is allot of metal and leather to be hauling around all the time and it manifests as bottom of the rankings efficiency.

Lastly was range:

Again the Teslas own this category with their massive battery packs but they do it with brute force, sporting the lowest efficiency and highest mass in the group. The i3s are distributed in the middle of the rankings, depending on configuration, generally ahead of other vehicles that offer much less in other function and features. It will be enlightening to see where they actually fall out after EPA testing.


The i3 has clearly established a new bar in efficiency demonstrating what is possible with a clean sheet, purpose built design. It will likely not be challenged until more purpose built EVs are brought to market. It is not the smallest or the lightest or the least utilitarian or the slowest EV from the set the EPA has tested which one may expect from the vehicle with the highest efficiency. In fact quite the opposite – it also exhibits strong performance by other measures including acceleration and with its low curb weight, it should prove to be very nimble and has already been reported to possess great handling. All with seating for four adults and a cargo area. In my opinion it is the best all around performer due to all of these considerations, the most significant of which is its EV leading, carbon footprint reducing efficiency.

footnote yiiikes

Categories: BMW

Tags: ,

Leave a Reply

79 Comments on "BMW Builds The Most Efficient “Purpose Built” EV Yet – The i3"

newest oldest most voted

I guess we don’t talk much about efficiency because it just isn’t as important in and of itself. Lets think about it. If you have two cars and they have the same range, same performance, etc.. But one is more efficient than another, then you might save a few cents of electricity. When talking about gas cars, efficiency can mean the difference of a hundred dollars or more of gas money each month. But when talking about an EV, there really isn’t that much difference is operating cost.

What we really care about are things like range, performance, etc.. Range probably being the most important. If higher efficiency helps us get more range, then great. But there are other ways to approach it too, like a larger battery pack or a range extender. But ultimately, the thing people want is range. How it is achieved seems to be less important to the consumer.

Very much true

This is a very good point and I applaud your efforts. That said, the data you present is a bit skewed. One cannot take the usable battery capacity, like you did with the i3, and the rated capacity, like you did with the LEAF S, Fiat 500e, Honda Fit EV and many others. Since most manufacturers don’t disclose the usable capacity, a more uniform and easily accessible metric is the rated capacity. I would recommend that it be used consistently for all vehicles. For the i3, it’s 21.6 kWh. Additionally, while the EPA range is a good metric, the protocol can vary a bit depending on the availability of 80% charging and the manufacturer recommendations to consumers in that regard. The LEAF, for example, archives 84 miles EPA rating on a full charge, but this was averaged with the 66 miles it achieves on an 80% charge. I would recommend that the EPA range on a full charge was used for the LEAF. The Toyota RAV4 EV is similarly handicapped. It’s worth noting that the EPA determines an MPGe rating, which reflects wall-to-wheels efficiency for all vehicles. It would have been interesting to reference those numbers for all vehicles listed… Read more »

This entire analysis appears to hinge on *projected/unconfirmed* EPA numbers for the i3.

Does the author have inside information? If not, what is the basis of his projections?

The i3 was rated to have less range than the LEAF by the NEDC tests. EPA results will be lower and more realistic for both vehicles, but why should we expect the i3 to leapfrog the LEAF?

For the EVs sold in Europe, a kWh per 100 kilometers metric based on the NEDC test cycle is a a available. For comparison, the i3 achieves 13.5 kWh/100 kWh and the 2013 LEAF achieves 15 kWh/100 km. That’s a difference of 10% in favor of the i3. This is roughly in line with the values projected by Darren in his article.

The article gives the i3 16.67% more range than the LEAF for a battery size 21.67% smaller.

Both numbers are off, so it’s a miracle if the LEAF only registered a more realistic 10% less efficient.

The rex version is going to be less efficient in cold climates due to lack of a heat pump. This will show up to some extent in the EPA 5-cycle test.

Yes, as I pointed put above, the EPA range figure gor the LEAF was determined as an average of 80% and 100% charging. The range achieved on a full charge on the EPA test cycle was 84 miles. This is a well-known and documented fact. The rated capacity of the i3 is 21.6 kWh, which is 10% less than the LEAF. While we can only guess at the EPA range of the i3, I believe that it will come in low 80s. While my personal pick is 83 miles, I hope we can agree that it will be somewhere between 80 and 90 miles.

It will be disappointing if the EPA 100% charge range for the i3 is not > 80 miles, but I am not convinced the 80/100 average will come in over75 miles.

I do like the i3’s less weight being less stressful on roads and bridges, though.

I don’t believe that the EPA will use an average of 80% and 100% charging to determine the range figure for the i3. Much like the ActiveE, the i3 might not even have an 80% charge setting. Please note that the EPA will only apply this protocol if the 80% charge is deemed to be standard and recommended S such to the consumer. Everything I have heard and every conversation I had hints at a figure between 80 and 90 miles, and I’m confident that this is what we will see.

Even the worst efficient BEV is better than ICE if we talk about CO2 trail.
So cheaper BEV -> more people will get rid of the ICE -> less CO2
Expensive BEV even if it’s super efficient – >very few people will by it. Many will keep driving ICE -> more CO2 emissions.

Very enjoyable article. Thanks for taking the time to post. Though I do think inefficiencies matter, I don’t think they are the most important, at least not at this stage of the game. I have utility company management tell me on regular occasion that my solar panels do not make sense because the efficiency isn’t there yet. Now if I and others heeded their warning, then the solar industry would not be moving forward. By simply buying more panels now I have helped the solar industry move forward and helped future generations as well as lower my carbon foot print. The same applies to the EV industry. It is more important to get the numbers up than to argue that you should buy an ICE if you don’t buy the most efficient model available (not a good argument). Right now, we simply need EVs out there and hats off to the battery bloated Model S and the desire it has created to reach deeper into the luxury market. Also to all the PHEVs and EREVs out there as well, for they too move us closer to the electric highway. In the end, efficiency matters and starting the conversation is worthy.… Read more »

The metric that matters is wall-to-wheels energy consumption at a number of constant speeds, uncontaminated by driver, undulating road surface, wind, nor usage of auxiliary power consumers such as heater and aircon. Heater and aircon usage, especially, should NOT be a part of the energy consumption efficiency metric – heck, in some cars, running the heater is equivalent to driving at 45mph. Aerodynamic drag, by far, plays the greatest role in determining range at highway speeds, and I am surprised that vehicle manufacturers are not jumping onto this ‘free’ range extender. (Aptera, RIP). Most of us with EVs know how to play the ‘get me home’ game of simply slowing way down and turning off the heater with dramatic range-extending results. FWIW, I think it is absurd of the EPA to consider the Leaf’s 80% charging limit in their published range numbers – if you want to take a longer trip, then you can always charge to 100%.

Went to the store to buy LED bulbs on Saturday. Drove one mile a minute down the interstate. Was passed by everybody, including all the Prius drivers.

If people cared about efficiency, and CO2 footprint, we would have been driving 100 mpg cars for at least the last decade. There is no secret formula. Throw out all the convenience, and infotainment junk to save weight, reduce frontal area, and make it round in the front and pointy in the back. Set an electronically enforced 60 mph speed limit. Leave something for our grandkids.

Dr. Kenneth Noisewater

Driving a quick EV is like having delicious cheesecake with 0 calories.

Darren, thanks for bringing in again the perspective of energy efficiency, and footprint in general. I’ve heard that BMW also took care to lower the production footprint, which for BEVs will eventually (as the grid cleans up) end up being the largest component of lifetime footprint.

That said, I second those who doubt that the i3’s EPA ranges will be all that impressive. In particular, the REx range might be a real dud.

That said… it’s a BMW… it will sell!

Assaf, I’m confident that the REx range and performance will be just fine. That said, I agree that the BEV range was a bit better.

Thanks so much Darren for doing this analysis. Very interesting, and very helpful to those of us who are seriously considering the i3.

this does nothing for me

The math in this “analysis” is fundamentally flawed. You simply cannot take the battery capacity and divide by the range to get the kWh/mile figure as the author did for the i3 in the first table. The most glaring problem with this is the charger efficiency. The EPA numbers for kWh/100 mile and MPGe are calculated using kWh from the wall, not kWh from the battery, so the efficiency of the on-board charger always comes into the picture. Also, as previously noted, several of the cars have had their EPA range figures averaged down because of different charge capacities made available to the driver.

My LEAF dashboard is currently indicating an average of 4.5 miles/kWh (most of my driving is slower city miles with a lot of regen). Am I correct is using this number to calculate MPGe since it is given after the ‘charging cycle’?

The MPGe figure the EPA uses represents wall-to-wheels efficiency. The number we see on the dashboard is battery-to-wheels efficiency. They differ by about 15 to 20%.

This comment of yours makes no sense…(I am only analyzing BEVs so I just don’t want to look at MPGe, too much ICE in it for me). The EPA’s MPGe formula was specifically designed so that the general public can better understand an EV’s efficiency in ‘miles per gallon equivalency’ rather than ‘miles per kWh’. Take the i3’s best number of 20.9 kWh/100miles and convert that to 209 Watts per mile then divide 33,705 BTU’s by 209 = 161 MPGe. It would only take a few minutes to add an MPGe column to your tables and make them far more comprehendible in general.