2017 BMW i3 REx Range Put To The Test Against 2014 i3 REx


255 miles of combined range? This range estimator is probable a little more optimistic than real life, but I definitely believe 200 miles is possible with the new 2017 i3 REx.

255 miles of combined range? This range estimator is probable a little more optimistic than real life, but I definitely believe 200 miles is possible with the new 2017 i3 REx.

The 2014 (60Ah) i3 REx vs The 2017 (94Ah) i3 REx

What a difference three years makes.  I was able to secure one of the first 2017 BMW i3 REx cars that made its way into US dealer inventory, compliments of Chris Chang, Sales Manager at BMW of Bloomfield here in New Jersey. The vehicle is mostly the same as my 2014 i3 REx, the one big exception is it has the new 94 Ah battery cells, which increase the overall battery capacity from 21.6 kWh to 33 kWh without increasing its physical size. That was necessary, since this isn’t a redesigned i3, so the battery modules had to fit in the existing battery tray.


The 2017 i3 REx in Fluid Black next to my “Moloughney Red” wrapped 2014 i3 REx

As much as I wanted to check out the moonroof option that this car had (finally the moonroof is available in the US!), there is no denying the single most important improvement I was interested in was to find out how much more range the new model has. The EPA range rating for the 2017 i3 REx is 97 miles per charge, and BMW claims 180 miles of total range when combined with the added miles from the range extender. The full 2.4 gallons of gas is now available for the REx. Previously it was software limited to only 1.9 gallons so the vehicle would qualify as a BEVx vehicle. My 2014 i3 REx has an EPA rating of 72 miles per charge, and BMW claimed a total range of 150 miles including the range extender miles. So the new i3 REx should offer about 35% more all-electric range, if the EPA test results are accurate. One thing to note is the auto manufacturers do the range testing in house, and reports it to the EPA. I think many people are under the assumption that the EPA tests the cars, and they do not. Manufacturers have been known to “massage” these numbers to fit their needs.

Range Testing

I wanted to perform three tests. The first was to fully charge the car and drive it easy. I didn’t hypermile, but I took it a little easier than I usually drive. It was 83 degrees, which is favorable for good range, but I did have the A/C on the entire time. I drove in Comfort Mode because that’s pretty much the only mode I ever drive in. I took a combination of highway and secondary roads and basically drove the speed limit with moderate acceleration from stops.


After 100 miles of driving, the car still had 26% state of charge and was estimating an additional 37 miles available. I’ve driven my i3 long enough to know how far it can go, within a couple miles, and I’m sure if I were driving my i3 in those same conditions it would have gone about 72 to 76 miles before the range extender would have needed to turn on. This new i3 REx easily beat the 35% range increase expected by the EPA range rating. In fact, based on these results I think it would be hard for me to get less than 100 miles per charge even if I tried. So that’s what I did for the next test.

This time I was going to drive it harder. Not Autocross hard mind you, but I’d punch it from all the stops, drive 75 – 80 mph on the highway and not concern myself with using the regenerative braking to their fullest advantage. Basically, I’d drive like I was late to an important meeting. Halfway through, I realized my efforts weren’t making much of a difference. At 50% SOC I had driven 62 miles and the range estimator still showed 62 miles to go. I did noticed that the gas range estimate had dropped from 85 miles to 75 miles though, even without using any. That’s because my driving efficiency was much worse than it had been on the first 100 mile drive.


Seeing how I was still on my way to 120+ miles of range, I stepped up my assault on the tires, and really thrashed the car around a bit. It worked, and I further reduced my efficiency. I finished up this 100 mile trip with only 13.5% SOC and estimated 16 miles remaining. I was able to reduce the single charge range by 21 miles, but I couldn’t manage to get less than 100 miles of range, which was my goal. In my opinion this is great news.

Honestly, I don’t know how this car is rated at 97 miles per charge; that’s nearly impossible to attain unless it’s being operated in cold weather or perhaps being driven at a very high rate of speed. I’m sure once the winter months roll in and the temperature drops it won’t be hard to get less than 100 miles of electric range. However in moderate temperatures, I think most people will always be in triple digits. Based on the experience with my car, I’m guessing this new i3 REx will probably average about 85 to 90 miles of all electric range in the winter. My car only averages about 60 to 65 miles of electric range when the temperatures are below 30 degrees Farenheight, therefore 85 to 90 miles sounds about right for this new, longer range model.

Even with trying to get less than 100 miles, I still managed 100 plus an estimated 16 miles remaining.

Even with trying to get less than 100 miles, I still managed 100 plus an estimated 16 miles remaining.

The REx Test

The final test was to see if the range extender performance was any different. Much has been made over the fact that the i3 REx can enter Reduced Power mode, and slow down under certain strenuous driving conditions. So I depleted the battery, drove it for 50 miles and made sure to take it up some hill climbs at highway speeds. The first thing I noticed is the range extender operates the exact same way as it always has. It doesn’t turn on until the battery state of charge reaches 6.5%.

The “Hold State of Charge” option is still disabled here in the US, so if you want that feature, it will still have to be unlocked by coding the vehicle, as before. There was some speculation that the automatic turn on point of the REx might be at a higher SOC with the new model, but I can confirm that’s not the case. However, there were two observations that I noticed that were positive.

First, the range extender seemed quieter from inside the cabin. In my car, the REx motor is pretty quiet and unnoticeable until it kicks into it’s highest output mode. At that point you can definitely hear the scooter engine revving up high from underneath the rear seats. It’s kinda like your being chased by lawnmower. On long highway trips it will operate at its highest level for most of the journey and the noise is noticeable. I’ll usually turn the radio up a notch to cancel it out. With this new car, driving at a constant 75 mph to 80 mph the motor seemed much quieter than it does on mine. My wife was with me for part of this test and she also noticed. She actually asked me if the REx was even running. It seems to me that BMW improved the REx soundproofing. It does sound just as loud as before from outside the vehicle, but it’s definitely quieter on the inside.


Secondly, (and I’ve reached out to BMW for confirmation on this but haven’t received a response yet) it does seem like the REx motor has been tuned for a slightly higher output. I took the vehicle on highway roads that I drive on regularly, and have on occasion done so when the REx was operating. The range extender was able to hold the state of charge higher, and under more strenuous driving conditions than my 2014 REx can.

There’s one particular long incline that I drive every day. With my car, if I start at the bottom with 6% SOC and drive 70 mph up to the top I’ll deplete the battery to about 2.5%. I did this same test with the 2017 car and I reached the top of the climb with 5% SOC remaining. I repeated the climb with the same results. I also noticed that I could drive at about 75 mph on flat ground and maintain the 6% SOC. My car can maintain the SOC on flat ground with a constant 70 – 72 mph, but not any higher or the charge will slowly deplete.

I know the 6.5% buffer is now larger, because it’s holding 6.5% of 30 kWh instead of 6.5% of about 19 kWh, so that extra energy is definitely helping, but to me it appears that the REx motor has a higher output for the 2017 model. The REx motor in my car is rated at maximum power output of 28 kW. I wouldn’t be surprised it we find out the power has been increased to about 33 kW, but I don’t have any official confirmation on that. I’m just going on what I’ve experienced with the previous REx cars and how this new one compared to it. Another hint that I may be correct is the REx is now rated at 35 mpg, down from the 39 mpg which the previous models were rated at. I don’t think the extra 170 lbs alone would cause a loss of 4 mpg. I believe it working harder now to produce more energy, which was I’m guessing was achieved through a software adjustment.

After driving 42 miles on the highway I still had 70.5% SOC and an estimates 93 miles or range remaining. My 2014 i3 REx doesn't even go 93 miles per charge! The range of the 2017 is a substantially greater than previous i3s, even more than the EPA rating would seem to advertise.

After driving 42 miles on the highway I still had 70.5% SOC and an estimates 93 miles or range remaining. My 2014 i3 REx doesn’t even go 93 miles per charge! The range of the 2017 is a substantially greater than previous i3s, even more than the EPA rating would seem to advertise.

Faster Charging With A New Profile

Previous model year i3s were capable of charging at 30amps which, at 240 volts, gave a maximum draw of 7.2 kW. The new i3s can accept 32 amps which translates to 7.4 kW. Not a huge difference, but it can help if you’re waiting for the car to charge to a certain SOC so you can unplug and drive. I should note that most public charging stations are limited to 30 amps, so it won’t make a difference on those units. However at home, I have charging stations that can deliver 32 amps so I was able to monitor the difference. My car usually accepts 7.1 to 7.2 kW (depending on the voltage supply) but this new i3 was consistently drawing 7.3 kW to 7.4 kW, so I can confirm the onboard charger upgrade.


The charging profile of my 2014 i3 REx is on the top, and the 2017 i3 Rex is on the bottom. 
Both charged from 6.5% to 100%. The 2014 car charges fully in about 3.5 hours and the 2017 in about 4.5 hours.

I did observe something interesting while monitoring the charging profile of the new i3. Instead of the charge rate gradually tapering off as the SOC reached 90%, and slowing down for the final 40 minutes of charging, this car took the maximum rate nearly right up to the end of the session. I charged it three times to monitor this and it behaved the same way all three times. I’ve never observed this on any other EV. Normally, the vehicle slows down the charging rate considerably as it approaches the end of the session to slowly balance the cells. This takes place once the vehicle is over 90% and the final 5% to 10% of charging takes much longer than charging at lower SOC. That’s not happening with this vehicle. It only slows down slightly, and only for a couple minutes at the very end. The charging rate doesn’t gradually lower until it shuts off, it more closely resembles falling off a cliff. Interesting.

This Is Fully Opened

This Is Fully Opened

Finally, a Moonroof

The moonroof is a new option for the US. It’s been available all along for i3s outside of North America, and now it’s available here also. The moonroof is a $1,000 option and is a split version, having two openings separated by a solid center section. Each opening has its own manually-operated sunscreen, but the moonroof itself is one piece, and slides back with a push of a button. However it only opens about eight inches, slightly more than half of the actual opening in the roof. It’s not even large enough to stick your head out of it – not that you would want to do that; but the point is, it’s a small opening.

The moonroof does accomplish two things, though. It allows more light in the cabin, giving the feeling of it being more open. It also allows you to eliminate side window buffeting by simply tilting the moonroof open.

While the "Batt.Kapa.Max" isn't an exact measurement of the available capacity, but it is very close. Close enough to prove there's much more than the 27 kWh that BMW claims is available.

While the “Batt.Kapa.Max” isn’t an exact measurement of the available capacity, but it is very close. Close enough to prove there’s much more than the 27 kWh that BMW claims is available.

Available Battery Capacity – Surprise!

BMW states that the new battery is 33 kWh, and 27 kWh of that is usable. That’s only 81.8%  of the total pack, much less than the ~90% they allowed to be accesses on the 60 Ah battery pack. When I read that I wondered if it was perhaps sign that the new 94 Ah cells were less tolerant to deep discharge than the 60 Ah cells were, so BMW was going to be conservative with them. So when I fully charged the battery after the first 100 mile test run, I checked the hidden diagnostic menu and to my surprise it was showing a full 30 kWh accessible. So BMW is allowing access to roughly 90% of the overall pack, just like they do with the 60 Ah cells. That explains the extra range I’ve witnessed but it doesn’t explain why BMW’s official stance is that there is only 27 kWh accessible. Perhaps it’s for battery capacity warranty claims?

My i3 before and after installing the H&R Sport springs. It dropped the car 1" in the front and .8" in the rear.

My i3 before and after installing the H&R Sport springs. It dropped the car 1″ in the front and .8″ in the rear.

Gained Some Pounds

The only negative I’ve found is that the new battery is heavier, and adds 170 lbs to the curb weight (3,064 lbs to 3,234 lbs). This does effect performance a bit. The car doesn’t feel quite as responsive as my 2014 does. Without testing the performance times, I’d say it’s probably close to a half a second slower from 0 to 60 mph. Handling didn’t seem quite as crisp as mine either, but that might not be this car’s fault. It has the 19″ turbine wheels, and my i3 the 20″ wheels with the sport tires, which are wider and have a larger contact patch. I also recently lowered my car with sport springs from H&R which have improved the handling, so it’s not fair to compare the handling to my car.

The other performance change I noticed is the regenerative braking seems to be blended in differently. When driving slowly, it seems pretty much the same as my car does. However at higher speeds the car will coast more when releasing the accelerator. The regenerative braking doesn’t initially come on as aggressively as is does on my car. It will get progressively stronger if you continue to coast, but initially upon releasing the accelerator, the car freewheels a lot more than previous versions do. I like this for highway driving, as freewheel coasting improves efficiency. If you slightly depress the friction brake pedal, the friction brakes aren’t used, instead the car used first uses only regenerative braking, until you depress the brake pedal harder.

Summing Up

After a couple days and driving over 300 miles I feel it’s safe to say that I believe most people will find the average usable range greater than the EPA rating of 97 miles per charge. I almost wonder if BMW purposely underestimated the range a bit in an effort to under-promise and over-deliver. On my 2014 i3 REx, I’ve found the range to be pretty close to the EPA rated range of 72 miles per charge. I do average a few miles more than that during the warmer months, and about 10 miles less per charge during the winter when it’s cold.

But this new i3 has unexpectedly trounced the EPA range rating by a healthy margin. I think most people should average well over 100 miles of pure electric range on these vehicles. The range increase will undoubtedly push some people deciding on whether to go BEV or REx into the BEV camp. I know if I were buying one today I’d go BEV also. Getting this kind of range with the REx, I’m certain 125 to 140 miles per charge would be easy to attain with the 2017 BEV. That, combined with the ever increasing CCS DC fast charge networks, would really be all I need for all my driving needs.


Thanks again to Chris Chang and BMW of Bloomfield for providing me with the use of this car for three days of testing.

Categories: BMW

Tags: ,

Leave a Reply

49 Comments on "2017 BMW i3 REx Range Put To The Test Against 2014 i3 REx"

newest oldest most voted

Can someone explain how the software limited gas tank works? Does the range extender ICE leave you stranded on the side of the road with half a gallon of fuel still inside?

Great article! Thank you!

Nice write up. Great to see BMW is ironing out the kinks.

Tom, … It sounds like with “hold state of charge” enabled and a slightly bigger gas tank (5 gallons?) the i3 REX would be fine for extended road trips. Do you agree?

Funny how EVs tend to achieve much better range than the EPA rating while ICE cars tend to get poorer mileage, without careful driving. I average about 52 per charge in my 2012 Volr, so I am beating the GM/EPA estimate by 50%. I can do a bit better than the EPA rating in my wife’s hybrid, but maybe only 10-15%.

I think EPA strike the right balance. An ICE doesn’t loose winter range, like BEV. Tom admitted he’d expect 80-85 winter miles, from the new i3. I think EPA is appropriately factoring this in, if not still under-warning customers of winter’s effect on batteries.

The effect of winter is so great on BEVs, I think it could be argued their range should be represented regionally, on the Monroney. The counter to this is when range reaches 200, or 238 miles, warming the cabin becomes a smaller dent in the car’s overall range. I’ve gotten 25-30 miles from the Volt, multiple times. There, it counts.

Per the “An ICE doesn’t loose winter range, like BEV.”, well – that might seem to be true since the average car has the equivalent of a BEV with some 240 – 350 kWh or more of ‘energy’ in Gallons of Gas, so the loss is not as noticeable , however – it you had just said – “An ICE doesn’t loose winter range.” You would have actually been TOTALLY WRONG, since – for example – a COLD ICE in temps of -10 to -20 F, would take a lot longer to warm up to operating range, and in that time it would be burning some 3-4 X as much Fuel per hour, and if you are commuting short distance to and from work, it has that problem, at least twice per day, which definitely DOES cause a loss of range, which can be as much as 20-30%! Using ScanGauge – https://www.scangauge.com/ with displays showing Fuel burn by Hour settings on one of the displays, along with Water temps, you could graph the values, but since I don’t know if it has a Data logger, you would have to resort to various other was to capture the data, but it… Read more »

Did you measure the fuel consumption when using the REx? How does it compare to the 2014 MY and to the EPA numbers?

Good question. Good article but then left us hanging on how far the gas will take you.

Nice review- thank you. How much weight do you think the sunroof adds?

I am going to guess that by understating battery capacity and officially claiming only 27kwh, that gives them in reality more than 6% left on the battery when the ReX kicks in and helps them keep the power up in inclines to get rid of that ‘feature’. So they are understating range then thus circumventing the goofy CARB rule

Since BMW proposed the rule, it’s at least partially a goofy BMW rule.

I still don’t think CARB should have made the BEVx category eligible for white stickers. The rules made the car worse, and make the white stickers less simple.

Cell balance time may depend on their state of charge and imbalance. In new cars with new cells, the time may not be long. After few years of use, it may (or may not) get longer for cell balancing.

As for L2 charging at full rate right up to the end, SparkEV with 3.3kW would do that since DCFC at 99% was about 9kW rate when relatively new. Seeing how some Leaf at DCFC was 2kW with 50 kW charger while others at 5 kW at same SoC, last few bits of charging will be highly variable.

Thank you Tom for a very thorough test and excellent presentation.

Seems like a solid improvement for the i3.

Your range news strengthen my impression that with the larger battery, the BEV version becomes the more attractive one.

Extrapolating from your range data, it sounds like with the BEV (which is rated at 18% more range than the REx) in reasonable weather, you can easily squeeze out close to 150 miles, maybe even more. Paired with a CCS port, that should suffice for a good chunk of road trips with minimal hassle.

“…the BEV version becomes the more attractive one.”
Especially with the heat pump for cabin comfort in cold weather. It will be interesting to see how it does with the Bolt EV/Ampera-e once it is on the market.

*you’re being chased by lawnmower.

Great comparison! A quieter and more powerful REx is really a good improvement, alas at the expense of efficiency. Wish they included a Mountain mode if not a full Hold mode.

I just don’t understand why BMW doesn’t make a stretched version of this car. By adding one row of cells to the battery and extending the wheelbase by the same amount they will have a very roomy car with about 150 miles of range at a very low development cost.

I thought the same thing when I got mine a year ago. I never intended on having people in the back sear very often – BUT – my son is 6’3″ and fits in the back seat comfortably. It’s interesting to note that he can’t fit comfortably in the driver’s seat of a Volt with the seat all the way back.

Tom, really great write up. I’ve been reading your columns ever since the Active E days. Lots of really great info in here. Makes me want to upgrade from my 15 BEV to a 2017 BEV

Great review!

Thanks for always thorough writing full of good information.

Thank you for an excellent review. You have given me the information I have been seeking, which I couldn’t find anywhere else. This car will now meet my needs, so I’m going to order one.

Very Nice Review!

However….I still think the I3 should have been a 100+ mile EV to start with the update to 200+ miles. That capacity should have been built into the brand new architecture from the beginning.

Myself, I would not want gasoline anywhere near my EV. And won’t that engine require tune-ups, oil changes, and the I3 Rex would also need emissions testing? Right?

With the battery capacity upgrades for the Leaf, and Focus Electric, along with the new Bolt, the most expensive, less EV range i3 and i3 Rex get pushed to the bottom of the heap.

There’s a strong case to be made for the i3 Rex.

Batteries are going to age and at some point (8, 10, 12 years?) are going to need replaced.

I’m not sure carrying around an additional 33 kwh of battery is a better idea. The Rex won’t get many hours put on it per year. Properly cared for, I could see that engine lasting 30 years or more, … easily. It’s hardly ever going to need a tune up and 3 quarts of oil per year isn’t a big deal.

With nearly rust free construction, I could see the i3 Rex (and it’s original ICE extender) lasting decades.

IOW, when the time comes, I think a $6,000 to $7,000 -ish battery replacement will still make economic sense. But if you have to spend $12,000 to $14,000 ?? That’s likely going to send the car to the scrap heap.

For me, in New Orleans, this reinforces my decision to buy a used 2014/15 i3 REx when the price drops late next year (target $15k to $17k). Hack the software to get 8 l fuel capacity.

Less than 10% of my trips are more than 10 miles round trip. Extra kWh are useless to me. 32 F is cold weather, so range impact is limited.

I do need REx for the once every 6 years hurricane evacuation. Pack 10 or 20 liters of gas in military jerry can. 39 mpg will get me a bit further than 35 mpg. Gas stations should be open within 200 miles. Refill gas tank every 100 miles after that and recharge when possible.

I love stories like this. It goes right down into the details.

But the end sum of the 6.5% controversy still seems to be to hack the i3 to give it the European battery hold mode.

So I’m wondering if there is any real difference between a hacked old i3, and a hacked new i3 when it comes to keeping it from going into crawl mode climbing hills?

Tom – How is the ride quality with the extra weight? As you know, a lot of owners report darty freeway handling and a choppy ride (small wheelbase, somewhat firm dampers, etc.) from the i3. Has the weight made that better? Worse? No material difference?

I am only seeing 49-50 miles of max range from the REX on my 2015 REX… these numbers are WAY higher. (I’m coded to see the additional tank capacity, ie full capacity too). Anyone know why? Is this just my driving style? Generally Battery range is 60-70 or so when fully charged.