Friday, February 10, 2017

SparkEV is the most efficient car in world history

In previous blog post, I analyzed SparkEV acceleration performance. It was shown to have 0 to 60 MPH time of 7.2 seconds as claimed by Chevy, making it the quickest accelerating car in the world that cost under $20K.

One would think that such quick accelerating car would have awful efficiency. After all, conventional wisdom is that quick cars like Ferrari and Porsche get about 10 miles per gallon while one of the slowest accelerating cars on the road like Mitsubishi Mirage get about 40 miles per gallon. Even among small cars, the “powerful” Chevy SparkGas with about 100 HP gets about 35 miles per gallon while 70 HP Mirage gets 5 miles more per gallon.

But upturning this conventional wisdom is SparkEV. I find that SparkEV as the quickest car in the world under $20K is also the most efficient car in history!

EPA says otherwise, buddy!

One could search the Internet and see that older model BMW i3 with 22 kWh battery was rated by EPA as 124 MPGe while SparkEV was only rated for 119 MPGe (second place, tied with Bolt).  However, the new BMW i3 with 33 kWh battery is only rated 118 MPGe, making it less efficient than SparkEV.

But we know better than to take EPA’s word for it. After all, EPA rated 2014 SparkEV range the same as 2015 that has smaller battery while Tony Williams range test showed that both 2014 and 2015 have greater range than EPA rating when driven at 62 MPH. Since 2014 has bigger battery, Tony’s test also showed more range for 2014 than 2015. EPA’s average test speed for MPGe determination is far less than 62 MPH, so Tony’s real-world test should’ve resulted in less range than EPA rating, which it clearly was not. You can’t trust the EPA numbers when it comes to real-world efficiency.

In this blog post, I’ll explore why SparkEV is the most efficient production car in the world as well as in history.

Average efficiency over 16K miles

To start off with, my 2015 SparkEV has about 16K miles on the odometer and average efficiency in almost 2 years is 5.3 mi/kWh. At 33.7 kWh/gal of gasoline, that’s equivalent to 179 MPGe! I haven’t reset the trip meter since getting the car, and below is a screen shot of current trip meter that shows average miles per kWh.

Some forum posters claim they have over 6 mi/kWh after over 10K miles of driving. I would've said that's impossible nonsense that defy the laws of Physics if I didn't experience this for myself. SparkEV routinely get over 6 mi/kWh (202 MPGe) when driven in traffic clogged city.

Some might claim that I’m a hypermiler, which isn’t true. I drive mostly in rural roads with average speed of about 45 MPH. Still, there’s the question of my driving being too conservative. A more standard test is to set the cruise control at some speed without wind and roughly similar elevation and record the efficiency.

There are several such tests: one by Digital trends web site at 24 MPH, one recorded by Tony Williams at 62 MPH (100 kph), and another by me at 70 MPH. While 24 MPH and 62 MPH are oddball speeds, 70 MPH is something that is done regularly on the freeway, especially since roads with speed limit of 65 MPH have people driving at 70 MPH or more.

24 MPH efficiency

24 MPH resulted in 7.2 mi/kWh with 2014 SparkEV. 7.2 mi/kWh is equivalent to 243 MPGe! Unfortunately, there aren’t test done at this speed with other cars to compare this number.

62 MPH (100 kph) efficiency

Tony Williams performed range tests for various EV at 62 MPH (100 kph). His test results can be found at the following link, and below that is the summary of results sorted in order of efficiency. As you can see, SparkEV clearly dominates in terms of efficiency in the real world driving at 62 MPH.

pre 2017 Car @ 62 MPHmi/kWhMPGe
BMW i34.7158.4
BMW i3 Rex4.6155.0
VW eGolf4.1138.2
Nissan Leaf4134.8
Kia Soul EV4134.8

But do I trust some third party test result? Considering Tony Williams runs the best EV after market product company in the world, I have high degree of confidence that he ran the tests properly.

Public service announcement

As a side note for those with Rav4EV or Tesla Roadster who wish to use DCFC, you can contact Tony’s company for installing Jdemo which allows you to use Chademo DCFC. If you have other EV without DCFC, watch for his company to see if they’ll announce DCFC product for you. Your “toy car without DCFC” that could only muster about 30 miles from home could drive literally thousand miles in a day with DCFC, just like real cars.

And no, I don’t get commission from Tony. I like endorsing great products, just like how I endorsed ev-vin’s lease blog. If you’re interested in leasing an EV, check out ev-vin’s blog.

70 MPH efficiency

I drove at 70 MPH, and found the efficiency to be 4.4 mi/kWh (148 MPGe). The test conditions were:

1. Charged using DCFC to 80%. It showed 73 miles remaining. Elevation at starting DCFC was 452 feet. There were two dogs and about 50 lbs of gear in the car plus the driver for a total of about 400 lb additional weight.

2. After charging to 80%, drove about 1 mile to freeway at about 30 MPH average speed (couple of traffic lights, all green).

3. Set the cruise control at 70 MPH in freeway. There wasn’t much traffic, which allowed this speed all the way, although there were few instances that I had to accelerate beyond 70 MPH for short time to pass slow semi-trucks.

4. After driving about 52 miles in freeway, I pull off the freeway.

5. Drove about 1 mile at average speed of about 40 MPH to DCFC. Elevation at ending DCFC was 698 feet, about 250 ft elevation gain.

6. Upon arrival at DCFC, car reported 8 miles remaining, and DCFC reported 10% battery remaining. Most importantly, it reported 4.4 mi/kWh for 54.05 miles trip.

Yes folks. SparkEV driven for 70% of its battery at 70 MPH AND going up 250 ft in elevation resulted in 4.4 mi/kWh (148 MPGe). Sure, it had about 2 miles of lower speed from/to DCFC, but that’s more than compensated with elevation gain and sporadic speed-up to pass slow trucks.

But you might be wondering if that’s any good. After all, Bolt is rated the same EPA highway MPGe as SparkEV, and old BMW i3 was EPA rated even higher. Unlike 62 MPH test by Tony Williams, there is no one place that ran the tests at 70 MPH in the real world for various cars. Then we google some test results from various places and come up with some numbers. Below table shows the findings and the source of the numbers.

pre 2017 Car @ 70 MPHmi/kWhMPGedata source
Chevy Bolt4134.8
BMW i33.7124.7
Fiat 500e3.6121.3
Tesla S 70D3.429115.6
Nissan Leaf2.9298.4

Clearly, SparkEV is the most efficient car in the world!

Other cars and speeds

For Renault Zoe, which is not available in US, a forum post shows 55 MPH resulting in 4 mi/kWh.

SparkEV’s 4.4 mi/kWh at 70 MPH is even better than Zoe’s 4 mi/kWh at 55 MPH.

For Mitsubishi iMiev, a forum post shows 65-70 MPH result in 39 miles to low battery light and 45 miles to “turtle mode”.

iMiev has 16 kWh battery. Assuming only 14 kWh out of 16 kWh battery was used, 39 miles would be 2.79 mi/kWh. Even if you assume 45 miles, that still only 3.21 mi/kWh. (if you assume more battery used, that result in worse mi/kWh) Either way, they’re all worse than SparkEV’s 4.4 mi/kWh at constant 70 MPH.

Some caveats

How trustworthy are these number from various forum posts? I leave it up to the reader to do deeper research. But from what I found, SparkEV is clearly the most efficient car in the world. Because EV are far more efficient than cars using any other form of energy, and modern EV came on the scene just few years ago, SparkEV is the most efficient car in history!

Above discussion on efficiency was energy efficiency from battery-to-wheels point of view and assuming 33.7 kWh per gallon of gas in computing MPGe. It did not take into account charging loss. Charging efficiency using 120V is about 80%, 240V is about 85%, DCFC is about 94%. Since the 70 MPH test was done only using DCFC, we can analyze the actual energy from outlet-to-wheels rather than battery-to-wheels.

At destination DCFC in my 70 MPH test, SparkEV at 10% took 17 minutes 30 seconds and 13.07 kWh to get back to 80%. That is average power of 13.07/(17.5/60) = 44.8 kW. You might be wondering why it’s so low since SparkEV is capable of 48 kW to 80%, making it the world’s quickest charging EV.

Higher voltage as the battery is filled result in higher power if the current is the same (power = voltage times current). SparkEV pulls roughly the same current to 80%, so the peak power is higher while the average power would be less. In any case, close to 45 kW to 80% isn’t bad. Moving on.

54.05 miles taking 13.07 kWh is 4.14 mi/kWh.

The charging efficiency for the trip is 4.14 / 4.4 = 94%

To calculate outlet-to-wheels MPGe, one has to take 94% of battery-to-wheels figure. Even if we assume all other cars in 70 MPH test were from wall to wheels, which isn’t the case, none of them is more than 4 mi/kWh, a figure SparkEV easily beats with 4.14 mi/kWh.

Some gasbags might argue that electricity distribution is only 90% efficient and generation is only 30% efficient. While technically true in some rare instances (ie, peaker gas turbine generators), gas cars only consider efficiency from what’s in the gas tank, not what energy was used to explore, drill, ship, refine, military to invade / protect countries with oil etc. Rather than analyzing all the complex scenarios, we simply analyze what comes out from the energy stored in the car (battery to wheels).

What will beat SparkEV in efficiency?

The question that comes to mind is when will there be a car that is more efficient than SparkEV? Unfortunately, it doesn’t seem likely that will happen any time soon.

First is the question of weight. SparkEV is only 2866 lb while Leaf, Bolt, etc. are close to 3500 lb. Even the BMW i3 that uses carbon fiber to reduce the weight now has larger battery and weighs similar to SparkEV while Tesla S weighs close to 4000 lb. From rolling resistance point of view, SparkEV with small eco-tires will probably remain the most efficient EV in the world.

Second is the question of aerodynamics. While SparkEV has pretty awful drag coefficient at 0.324, it has very small frontal area. Combined, they contribute to small overall aerodynamic drag. While I haven’t investigated this further, the test result at 70 MPH shows that SparkEV performs quite well at highway speeds.

Third is the question of drive train efficiency. SparkEV is unique among EV in the gear ratio. All the other EV hover around 9 to 1 ratio (9 motor turns for 1 wheel turn). Even the Chevy Bolt hover around 7 to 1 ratio. SparkEV by comparison is about 3.17 to 1 ratio (2015+ has 3.71 to 1 ratio). That means the motor is turning less than half the speed of other EV for given road speed. While the motor efficiency is a complicated subject, slower turning tend to be more efficient when all other factors being equal since moving parts don’t have to drag around as much “fluid” (aka, air, lubricants, bearing friction).

A potential candidate that could beat SparkEV is Tesla model 3. Tesla 3 is supposed to have drag coefficient of about 0.21, but it’s also larger frontal area as well as heavier. It also has Tesla drive train that is much more powerful, which may not be as efficient as less powerful SparkEV drive train. Judging from Bolt that is more efficient than much less powerful Leaf, it seems GM engineers make the most efficient drive train even with more power. It is unlikely Tesla 3 will beat SparkEV in efficiency.

Another candidate is Hyundai Ioniq electric version (not the hybrid/plug-in hybrid). Supposedly, that has the best EPA MPGe rating among any car at 136 MPGe and much less powerful motor than SparkEV. But I discussed EPA’s problem earlier in this blog post. Until someone performs a real world test when (if?) the car becomes available, we won’t know if it’ll be more efficient than SparkEV. I suspect SparkEV will be more efficient, just because I am thoroughly impressed with Chevy engineering.

So for the foreseeable future (until SparkEV 2.0?), SparkEV will reign supreme as the most efficient car in history.

Profile of history’s most efficient car

From the test results, it seems the 2015 SparkEV has the same efficiency curve of 2014 that I deduced in earlier blog post.

Below may be what the miles per kWh (or MPGe) profile of history’s most efficient car looks like. I challenge the world to beat that.

Edit: 2017-06-28

Normally, I don’t pay attention to car’s efficiency. But going for a long drive (125 miles each way) and traffic slower than the speed limit, I got bored and started doing mi/kWh math in my head.  The car shows MPH and kW of power being used, it’s simple matter of division to figure out mi/kWh (I think this is called mental masturbation, but I digress). I was mostly getting over 6 mi/kWh and often 8 or 9 mi/kWh! I was often under the speed limit but still about 45 MPH on average, not crawling at 20 MPH.

When I got to the DCFC station about 80 miles away after about 1 hour 45 minutes of driving (1.75 hours), the car reported 6.4 mi/kWh (216 MPGe battery to wheels)! Average speed was 79/1.75 = 45 MPH. The elevation change was about 100 ft down from the starting point, but that has little effect.  Basically, if one’s in light traffic, achieving such phenomenal efficiency with SparkEV can be expected. Below is the screen shot.

The energy used in this drive was 79/6.4 = 12.34 kWh.

Then on the way back, I was curious if I can achieve the same efficiency. Unfortunately, the starting DCFC was not the same as the ending DCFC location from previous drive. In addition, the traffic was moving quite well. There were some sections of stop-and-go, and other sections where the speed was over 65 MPH, but I generally kept the cruise control set to 60 MPH. I drove 65 miles taking 1 hour 15 minutes, average speed of 52 MPH. The efficiency showed 5.4 mi/kWh (182 MPGe battery to wheels)!

The energy used in this drive was 65/5.4 = 12.04 kWh

Then the total distance for recorded section of the trip was 79+65=144 miles taking 12.34+12.04=24.38 kWh. That results in average efficiency of 5.9 mi/kWh (199 MPGe battery to wheels)!

At $0.20/kWh in San Diego electric prices, 24.4 kWh is $4.88. Gas prices are about $2.75/gal, so the cost to drive 144 miles is about 1.77 gallons of gas, or 81 miles per gallon equivalent in terms of money out of pocket (MPGe$).

That’s the real-deal: SparkEV costs less to drive than any gas car, probably less than any EV, while being the quickest car in the world that cost under $20K when new. Now THAT is an engineering marvel.


  1. Sorry to burst your bubble, but the Spark EV is NOT the most efficient EV at 24mph. I have the best kWh efficiency in a 2013 LEAF for a 100 mile economy run (real world driving) at 8.5m/kWh that I did in 2013. Then, after everybody had finished, I continued to see how many more miles I could go and I went 155 miles total! Later, on Memorial Day, I drove 188 miles on one charge at 24mph on a regular street in the city with a 5% loss of capacity so I didn't even have a full battery capacity or I could have gone maybe 200 miles! I obtained over 10m/kWh! So I have the unofficial world record, not only for efficiency, but also for total miles on one charge in a 2013 Nissan LEAF. I have all the documentation for both records, but I'm sure Tony Williams will remember that I posted all that on the LEAF forum in 2013.

  2. Leaf has about 22 kWh usable from 24 kWh battery. SparkEV test by digital trends was for entire battery capacity, so your test of 155 miles out of 22 kWh would be only 7.0 mi/kWh compared to SparkEV’s 7.3 mi/kWh. Apples to apples, SparkEV is more efficient than Leaf at 24 MPH.

    Your claim of 188 miles and over 10 mi/kWh would mean usable battery capacity on Leaf would be less than 18.8 kWh. That is less than SparkEV battery. I doubt Leaf has smaller battery than SparkEV.

    Even if we assume that your Leaf battery has degraded by over 25%, which is probably very common among Leafs nowadays due to Leaf’s poor engineering, 188 miles at 24 MPH is almost 8 hours of “real world driving”. Where do you drive at 24 MPH for 8 hours without stopping for traffic lights and cops pulling you over? Even digital trends article describe issues with bio breaks. Your claim here is total BS.

    And if you’re talking anecdotes, I drove a same piece of road one way, then the other way immediately to discount the effect of wind. Due to traffic, average speed was about 20 MPH in each case (about 1.5 hours of driving total; you can guess this was LA). I got 12 mi/kWh one way and 8 mi/kWh the other way (average of 10 mi/kWh, real world driving including stopping). I don’t include such anecdotes, because had there been a tail wind, I could get infinite miles per kWh.

  3. I've been the happy owner of a 2015 spark EV for a month. Love it. I haven't reset the trip, and I don't know if the previous owner ever did. It had a little over 14,000 miles when I bought it .
    Warm climates are definitely going to see better results .I'm sure every EV sees a huge drain on the battery when using the heater. In Wisconsin, March into April still has temperatures hovering around freezing. In an attempt to see what this little electric hot rod is capable of in terms of mile per kw, I've been selectively using the heater.
    Long story short, without any reset I'm showing an average of 3.7 miles per kw. The car was shipped to me from NC, so I'm sure there was some heater use during the winter .
    With selective heater use the predicted range has been climbing . at purchase it only showed 59 for the middle prediction! Now I'm seeing close to 90 for a max. Curious to see how it will do once we finally warm up...first week of MLB with snow flurries anyone? In any case, call me biased but I loved your article .hopefully long term maintenance proves to be better than average...I already see how quickly I'm going to need front tires. But if that's the biggest issue then call it a win.

    1. I have about 28K miles on my 2015, and it's showing 5.4 mi/kWh. It used to show 5.5 mi/kWh, but I replaced the tires with non-eco tires (aka, normal tires). Of course, this depends on where/how you drive. But considering that none of my gas cars gave more than EPA rated MPG, the efficiency of SparkEV is phenomenal.

      Each drive session gives mi/kWh figure, so you can see how you're doing with spot checks. I saw 5.3 mi/kWh after 1 hour at 55 MPH with normal tires.