My 134,000 mile BMW i3 had only 48 miles of estimated range when I first test-drove it. “Oh boy, I think the battery is toast” I worried before asking the BMW dealer to do a battery health check. He refused. “You’ll have to buy it first or someone else will probably snag it. Only then will we check the battery,” the dealer told me. I took the risk, and the battery ended up being toast. Sadly, BMW’s battery warranty is just eight years, 100,000 miles; my new i3 is nine years old with 134,000 on the clock. So I was screwed; or so I thought.
If you’ve been following my i3 saga — my very first foray into the world of electric cars, and one that involved me, a man who tends to buy old Jeeps, falling in love with a little carbon-fiber car with an incredibly beautiful interior — then you know that the California Air Resources Board saved my ass. Specifically, it did so via this rule, which oddly only applies only to plug-in hybrids/PZEVs, and not to fully electric cars:
![Vidframe Min Top](https://images-stag.jazelc.com/uploads/theautopian-m2en/vidframe_min_top1.png)
![Vidframe Min Bottom](https://images-stag.jazelc.com/uploads/theautopian-m2en/vidframe_min_bottom1.png)
My i3 is equipped with a “Rex,” or “range extender.” It’s a little two-cylinder, 650cc gasoline motor built by Taiwanese company Kymco and normally found in a BMW C650 GT scooter. With the car’s 1.9-gallon tank, the engine is there to help you get to a charging station once you’ve driven all 72 miles of EPA-estimated all-electric distance. In my case, the motor had an even more important role: It made me eligible for California’s 10 year, 150,000 miles warranty, which the dealer honored.
![Screen Shot 2023 05 09 At 1.52.26 Pm](https://images-stag.jazelc.com/uploads/theautopian-m2en/Screen-Shot-2023-05-09-at-1.52.26-PM.png)
I will say I was skeptical at first. The dealer had initially told me it would take six to eight weeks for the new battery to show up from Bavaria, then it called me 10 days later to say the job was done. After my first, rather tame drive saw the car going 86.2 miles, I was convinced the battery had been replaced; but I have lots of questions still: Is this battery the same as the one that originally came in 2014 i3s? Is Samsung still making the same, older pouch cells? Are my batteries brand new or have they been reconditioned in some way? Surely they haven’t been sitting in a warehouse since 2014.
I likely won’t ever know the answer to these questions, but what I can do is test the battery and see the maximum number of miles it can propel my i3 on a single charge. So that’s what I did.
I’m not sure “hypermiling” is technically the right term here, in that I didn’t try to draft behind vehicles on the freeway, I didn’t tape up certain features or pump my tires up to a high PSI, and I didn’t focus hard on steady-state cruising. Plus, I did take a drive through a canyon road, which didn’t help things; also, the temperature was pretty high on day two, and then you could argue that spreading this across two days wasn’t great since on day two the battery had to condition itself for a second time to maximize its efficiency. But nevermind all that: This test is me driving as carefully as possible to maximize range from this mystery battery that a BMW dealer chucked into my i3 thanks to California’s consumer protection rules.
Driving with efficiency in mind in an EV is a little different than doing so in a gas car; gas cars love highways — their engine speeds are steady and low in high gear, they’re going a long distance over a short period of time, and they’re not wasting gas idling while not moving. EVs are oftentimes not most efficient on the highway; it takes a bit of time to get used to the fact that your most efficient route may be your most miserable one: traffic.
I kind of like this, since with a gas car your misery sitting in gridlock is joined by more misery at the fuel pump, whereas in an EV you can at least enjoy a little efficiency win while laying on your horn on the 405. And it makes sense why EVs do so well in town and so poorly on the freeway: Electric cars don’t have to idle, since their motors make lots of torque even when not rotating. I’m sure you’ve all heard that in car reviews before: “EVs make max torque at zero RPM.” But what does that mean? Well, it means that when that motor isn’t spinning at all, it can create a bunch of rotational force (this is technically not the right term, but for the layperson, that’s a good way to think of torque); imagine grabbing an electric motor and stopping it. Maybe you have a remote control car, you pick it up, and turn it on — imagine taking those spinning wheels and stopping them so they cannot rotate. You feel that force your fingers? Those wheels twisting your skin a bit as they try to rotate? That’s what it means by “electric motors make torque at zero RPM.” Your remote control car’s motor is not spinning — it’s at zero RPM – but you can feel that it’s making torque, trying to spin.
A gasoline car, at zero RPM, makes essentially zero torque. I say “essentially” because the starter motor does temporarily create torque in order to rotate the engine up to an RPM that allows its pistons to squeeze are (build compression) just enough to facilitate combustion, which pushes a piston down and gets the car idling to the point where the starter is no longer needed. Most modern cars do shut their engines off in the city, and use beefed-up starter motors to turn them on just before you need to take off from a light, but many have to sit at stop lights and idle — burning gas to travel no distance whatsoever. It’s pure waste, and electric cars don’t have to do it. At a stop light, an EV is using as much electricity as it would be while off, plus whatever is needed to run accessories and maybe your AC.
What’s more, in city driving, you’re often on your brakes. In a gas car, brakes are literally a mechanism that turns your car’s kinetic energy into friction/heat. That’s literally their job. In an EV, and especially in my i3, letting off the accelerator pedal slows the car down dramatically, but this deceleration isn’t a production of friction braking, it’s actually the resistance associated with the electric motor acting as a generator to recharge your batteries.
So in the city, EVs don’t idle, and they leverage stopping at lights to recharge the battery. What about on the highway? Well, on the highway, they deal with a lot of the same challenges as an ICE car — rolling resistance, aerodynamics, bearing friction, powertrain inefficiencies, etc. But the biggest one of all challenges is that second one: aerodynamic drag, which follows the equation: Drag equals one half times air density times velocity squared times drag coefficient times frontal area.
Notice that the velocity is a squared term, meaning your drag goes up with the square of your speed. So imagine going from 25 mph in the city to 75 on the highway — you’ve tripled your velocity, so your drag has gone up nine-fold! Just intuitively, you can imagine pushing a big car at 5 or 10 mph — maybe you’ve had to do so while trying to give someone a roll-start in their old VW Bus. That’s not too hard, right? Especially with the i3’s reasonably light weight and skinny tires, that motor isn’t doing much pushing the little carbon fiber car down the street at low speeds. Now imagine pushing that same car, but with a 75 mph wind gust blowing against you — that’s the same thing that the car has to deal with on the highway. It’s quite a lot harder to propel that car forward, and thus, it’s why the i3 tends to fall on its face when traveling at high speeds. (There are also implications associated with motor gearing, but that’s really less important here).
For these reasons, I kept the i3 in town. Obviously, because regenerative braking isn’t perfectly efficient, stop-and-go traffic is still less optimal than steady-state cruising at low speeds, and high temperatures are less optimal than low ones since the car’s active thermal management will have to work harder to keep the cells where they want to be (which is about the same temperature as humans — speaking of, all of us in the car were boiling when driving in the valley). Still, this was a good test of the i3’s capability.
The trip began with the i3’s Guess-O-Meter estimating a range of 98 miles in Eco Pro + mode, which dulls the car’s acceleration, limits its top speed to 56 MPH, and shuts off access to air conditioning or heated seats.
Somehow, I convinced my two friends Jeb and Giulia to join me on this rather hot adventure; I even put down towels so they wouldn’t sweat on my i3’s wool seats, which are extremely hard to clean. Notice how, after driving just 14 miles in the city, the computer recalculated its expected range: It expected the car to manage another 96 miles, for a total of 110 miles!
The range estimates on the Guess-O-Meter moved around a lot, as the car recalculated how it expected me to drive based on my pedal inputs. Going up a steep grade made it think I was going to continue driving like a jack-hole, so it dropped the range estimate. Coming back down the steep grade made it think I was going to keep driving like a grandpa, so the range estimate went up. The highest total range estimate we saw was 113 miles. My goal was to hit 100, so this seemed like decent margin.
I drove Jeb and Giulia around the valley a bit, then to Santa Monica on the coast, where, mercifully (especially for Giulia, who has no window to roll down — that’s right, the i3’s rears don’t go down!), the temperatures dropped.
From Santa Monica, I headed to Topanga Canyon, which maybe wasn’t the most efficient move, but I was careful. The i3 was a lot of fun to throw around turns; its tire are skinny, but it’s a lightweight(ish) car, and a BMW:
Eventually, as the distance climbed, and we entered our fourth hour trying to exhaust the i3’s battery (something that would have been very quick to do if it’s maximum efficiency occurred on the highway) our GoPro batteries started to die, as Chris — Galpin’s incredibly talented director — is explaining below:
Darkness fell, which gave us all a break from the terror of that bright yellow ball in the sky, but now I was running late for a commitment I had made, and Chris was growing tired, too. This was a Friday night; the weekend had begun, and we were driving around in traffic for hours on end with dying GoPro batteries and full memory cards!
Jeb and Giulia were kind and patient, Chris and I were the former but less so the latter. Here’s me pointing out how the expected mileage left increased from 37 to 38 while just sitting at a stop light. This was maddening:
Eventually, it was 8 P.M., I was in a bit of hot water regarding my commitment, and Chris was done. So he and talented cameraman Carlo both ditched us. Here they are speeding away, disgusted by the longevity of my BMW i3’s battery:
Jeb, Giulia, and I met up with the person to whom I’d made my commitment, and after apologizing, we had dinner. Then we went home, and I continued testing the following morning, when it was really, really hot near my Studio City Apartment:
As you can see, I started the day having put 89 miles on the battery pack — well above the EPA-rated 72 miles! But I wanted to hit 100, and with only 13 miles on the Guess-O-Meter, I all of a sudden had almost no margin.
So I drove slowly on my way to an EV car show at California State University at Northridge. It was 16 miles away, so I wasn’t expecting to make that. But 100 miles? Surely, right? I mean, the Guess-O-Meter had told me 113 at one point, and was regularly estimating over 105 pretty much all the previous day! Now I had an estimated 102 miles of overall range, and yet, no matter how slowly and carefully I drove, the estimated remaining range seemed to drop more quickly than the miles traveled rose.
My heart started beating fast. “Come on. How?! We had this!” I exclaimed, lightly pressing on the rightmost pedal with my big toe. “100 Miles was in the bag!” No matter what, every mile I drove took more than a mile away from the Guess-O-Meters range estimate. I tried slowing down, but nothing changed! Then my range dropped to four miles after having driven 95.8 miles. That’s a total of 99.8 miles. “No!!!” I continued exclaiming while staring at the Guess-O-Meter like a hawk. Then it happened:
The estimated remaining range dropped from four miles to zero. And in short order: “Popopopopopopo” the gasoline range-extender under my rear floor fired up. The test was over:
I’d driven 96.7 miles, or 24.7 miles more than the EPA’s rated all-electric range, and 10 miles more than I’d gotten out the battery in my previous test. Though I didn’t hit my target, it was a clear indicator that my battery was healthy, and that indeed, paying just $10,500 for this high-mileage i3 with a beautiful interior was the deal of the century. All thanks to one California emissions/consumer-protection law.
Why I Bought My Currently-Broken BMW i3: LA Was Making Me Fall Out Of Love With My Old Cars
I don’t get how anyone would have spent $40k+ on a car that can barely limp along for 80 miles with no AC.
Well then, I can tell you’ve never enjoyed an $88 empanada at a Miami F1 Race.
I’m more of a $69 nachos person. Those are nice.
Wow, this is a bit crazy to see! But then, epaulets ranges seem to be beatable. But 100 miles is nothing to scoff at- what’s the range like on these if you modify the extender to come on sooner?
Also, you no longer have the cheapest i3… I found a dealer with a Rex model for 8,990. 2014 with 120k. And your articles are making this more difficult to ignore than an improperly labeled taillight at a Torchinsky convention.
Epaulets ranges, you say?
Sometimes, I wonder why autocorrect would know that, but not EPA?
“Drag equals one half times air density times velocity times drag coefficient times frontal area.”
…. missing the squaring of the velocity, it is stated in the next sentence, “Notice that the velocity is a squared term..”, it should be but the but not in the verbal description given.
Not sure how I lost that term, but thank you! fixed!
Your open windows possibly created more drag?
He had 2 extra people in the car.
Are you practice dating?
A dinner date after hours of driving in the heat with the AC off sounds like David all right.
That probably shaves 2-3 miles of your range. 😛
You should remind people that the dealer battery health test on the original battery returned 30 miles, not the 48 on the guess-o-meter, even though they didn’t tell you how they tested it.
Bad battery packs are non-linear because they need to protect any weaker individual cells from being reversed during discharge. I saw this happen when I added a homemade 4-in series AA NiMH battery pack to my running headlamp 20 years ago – it’s like sticking that weak cell on a charger backwards. You don’t want to reverse charge Li-ion or risk it going boom. I imagine that’s why most EVs with bad battery packs throw codes and stop working, rather than allowing themselves to be able to be used at, say 30% original health.
Slightly disappointed from the start, that clip of DT in the shower but no spaghetti in sight…
So spaghetti is where you draw the line on disappointment? I think most of us were just disappointed that we had another clip of DT in the shower, period.
Partially motivated by this series (don’t judge), I finally pulled the trigger an updated my 2014 i3 to a 2017 i3 REx. Funny story…
The car was offered at a non BMW dealer and it was rather competitively priced. When I requested the Carfax, it quietly showed that a couple of months ago BMW had replaced “a” battery. I checked with the dealer and they shrugged it off. Yeah, of course, that thing that you sometimes replace while changing the oil, the battery. No big deal.
They had no clue that it could be “THE” battery (this same dealer had chargers in their parking lot that they did not know how to use).
So I bought the car and… bingo! Since I got it, I am getting a seemingly real world range of 125 miles. This not only is the estimated range for a brand new battery, it is the estimated range for the updated battery realeased 2020. A 2017 model should top off at 97 miles.
Either the car did indeed get a new high voltage battery, or it is the best preserved battery I have ever seen.
Also, I am suspecting that BMW is replacing damaged batteries from the first and second generations with refurbished ones from the next larger size, so they don’t have to keep an inventory of the older, smallest batteries. They would not be able to use a refurbished 94ah battery for a 94ah car if the range is slightly below nominal, but the customer would be totally satisfied if the car was a 60ah.
Didn’t 2017s come with 94aHs?
Correct, that was the first year. And the range that I am getting is almost what 120 Ah would get. Similar to yours getting almost 94 Ah range on a 60 Ah.
Good to see someone else motivated by this series to get an(other) i3. I’m 6 weeks into ownership of a 2017 full EV, and absolutely loving it.
My car is regularly getting 120-130 miles full range from my 94Ah battery (original battery, 42000 miles done) so you may just have a brand new 94Ah (which is still a excellent result!)
Very entertaining watching adult Ernie Douglas hypermile his car. I would gladly sit on a towel for you but major hissy fits would ensue over the air conditioning. Nothing better than having a 650CC backup, eh? Now, how about a top speed run?
Recently shopped around for a used Volt – would have really been nice to have been in a CARB state and gotten the extra 2 years / 50k miles…
Gave up and bought a 10 year old Nissan Frontier with a stick instead.
Respect. I like the Frontier. Stick is a little mushy, but I dig it.
The distance travelled is especially good considering that you had two passengers in the car for most of it, and it sat overnight which probably resulted in some of the dreaded “phantom drain” that EVs are prone to when sitting for hours. (These things in addition to the battery reconditioning already mentioned.) A few circumstance adjustments, and getting over 100 miles would have been pretty likely.
Agreed. I would bet 20 quatloos David would have busted 100 miles with ~300 lbs less mass aboard.
I’ll see your 20 quatloos and raise you 3 bars of gold-pressed latinum
Latinum is liquid at room temperature, so good luck with that
“Phantom Drain” is really only a thing on these modern always connected internet EV’s.
I owned a Spark EV for 5 years and it didn’t lose a single mile for as long as it was off. I actually parked it in my garage for 2 months and when I pulled it out it had EXACTLY the same charge as when parked.
My new Tesla though loses around 3-5 miles a day just sitting around.
That’s nothing compared to some of the newest and least mature brands like Rivian that can lose 10+ miles a day depending. I think they need some software refinement.
I’d still say you got lucky on a technicality David. Had you been anywhere but California, you would have been stuck with a $10000 bill. Do a post op in 10 years and let us know how that battery is then. This kind of touches on the problem I have with EVs and PHEVs. Unless the OEM or the dealer is willing to eat the bloat costs, I don’t want anything to do with them.
David is referring to Interstates as ‘the’.
How long before he starts saying hella? Are we already too late?
Hella is a NorCal thing. We don’t say hella in LA.
What was your MPGe or equivalent (mi/kWh – kWh/100 miles)?
The best I can get in my Polestar 2 is around 34 kWh/100 miles or 130 MPGe in my Chevy Volt. We need to standardize this lol
I like mi/kwh. That seems most similar to the MPG numbers we are used to. kWh/100 miles confuses me since less is better. MPGe is easy to calculate if you know mi/kwh, but it seems easier to skip the extra step and just use mi/kwh.
My Leaf gets around 4.9 mi/kwh.
Another vote for miles/kwh. My 2017 Fiat 500e averages right at 5 miles/kwh in flat-land mixed town/roadway (<60 mph) driving (and about 3 in mountain driving). With 28,800 miles on the clock, it still gets more than the factory estimate of 83 miles per full charge (right about 100 if I really baby it). The factory estimate for mixed driving is 3.3 miles/kwh so I’m pretty proud of little Blueberry 🙂
And another. Miles/kWh is simple, and what you want to know. You buy electricity by the kWh, so you want to know how far that takes you. The other metrics are a fudge.
My i3 is getting 3.8m/kWh with a pretty even mix of urban/country lanes/A roads and dual carriageways.
I charge the car either by our solar panels, or on cheap overnight tariff at 7.5pence/kWh, therefore at very worst, it’s costing me 2pence/mile when my old Golf diesel was at best 17p/mile.
I’m jealous of your solar panels. I hope to get solar in a few years, though.
Where I used to live, I could charge overnight for $0.01 per kwh. My daily 40 mile round trip commute cost $0.09 (I think that is ~6 pence?). EVs are dirt cheap to run.
We’ve only had the panels, and a home battery for a few months (electricity prices went stratospheric here last autumn, bills are about 4x what they were two years ago).
As soon as the panels were up and running it became clear an EV was an absolute no brainer, I had planned to wait until we’d paid off the loan for the panels, but the savings are compounded having both the PV system and the car, so committed to that too.
So David was committed ? it was only a matter of time
If you go back to 1955 and burn the Grays Sports Almanac, he gets Commended instead.
FWIW, I got 100 miles out of my original 2015 battery i3 once. Battery was a few years old at the time, too.
Definitely an excellent deal. Congratulations for driving rusty crap for decades to finally succeed when buying a new car. Now after reading about ICE in traffic wastes energy and turns it to heat if the ICE is a manual or even an auto Dylan in neutral has it been tested for affect?
I always got strange satisfaction in hypermiling my PHEV’s. It really made me a more efficient driver on my ICE vehicles too. I squeezed just shy of 50 out of my Volt once (rated at 37). I haven’t had the chance to do much with the 330e, but I’ve gotten 25 ish with the right conditions without trying too hard (rated at 20).
Eco Pro Plus is a bit different in the 330e. It still reduces the acceleration, but just reduces the HVAC. The cool thing it does is use the Nav destination and live traffic to determine the best place to use electric vs. Gas. It does it really well too.
You might be more efficient running the AC than rolling down the windows. One of my favorite features of my Volt was just being able to sit someplace with the AC on if I was working between site visits or something. It took almost nothing to run the AC (like 1kw for 2 hours).
I find EV hypermiling surprisingly entertaining. Preserving momentum and maximizing regenerative breaking can be a fun challenge. My personal best is 103 miles in my craptacular first generation Leaf. Advertised range was only 84 miles when new, and my battery has ~75% of original capacity.
It is nice to have a vehicle that can be entertaining while not breaking any traffic laws. I don’t think I annoy other drivers too much (no one has rammed me or shot me yet, so I can’t be that annoying I guess?).
I did it once at night when there were no gas stations open in my Mirage, the pump showed that I filled the tank with the tank capacity plus .3o litres, it may be fun but not when you HAVE to do it
I have had more than a few occasions where hypermiling was a necessity and not entertainment. I think my record was pulling in to my driveway with a state of charge of 2%. That was when I drove 20 miles from home before realizing I forgot to plug in the night before. I have always made it to my destination so far, though. This would be much less of a problem if public charging infrastructure didn’t suck.
It is, as is the slightly different game of seeing how far I can go without touching the brake, using regen alone.
I usually drive my Leaf in “B” rather than “D”. I can’t quite come to a complete stop without touching the brake, but it is close. My Leaf has 70,000 miles and I doubt the brakes will need to be replaced any time soon (if ever).
My Livewire allows custom settings for regen. I have mine set to 100%. I often don’t need the brakes to stop. My only concern is that someone will rear end me since I don’t think the brake light comes on with regen alone.
I don’t think the regen can be adjusted on the i3, and it does bring the car to a complete stop. My record so far is 52 miles, but I’m sure I can beat it, I only had to break as someone cut in front of me on a dual carriageway.
Get someone to follow you and check if your brake light comes on, I’d have thought it’d be like the i3 where gentle regen doesn’t, stepping off the accelerator suddenly does
How on brand would it be if the commitment was a first date? Was the trip to Santa Monica a way to freshen up for dinner, Baltic Sea style?
Captain Obvious speaking… you would have made the century if you didn’t have 300+ lbs of extra people in your car.
Also stick to the flats.
This video just convinces me more that PHEV is a better option with today’s tech. The i3, while nice for a pure city car, does not look very useful for highway driving at all. For some folks that may be all they need, but I think for most that’s not going to cut it.
Car dealers hate him!
One weird trick to get a new battery for your BEV!
I bet the dealer loves him they got paid. If i was a CA Dealer i would hunt all of the eligable warranty cars down buy them, install new batteries at manufacturer expense and sell as a EV with brand new batteries. And screw over buyers and use the warranty for my own benefit.
That’s not a bad idea. If dealers get paid by the manufacturer to do the work, you may as well track down every broken 9.5 year old i3 you can find.
I wonder if you could import i3s to California or if this somehow excludes vehicles brought in from other states? You could buy broken i3s out of state for half price. Possible arbitrage opportunity here.
My understanding is that it has to be sold in that state and also never registered outside that state. That was the case for the Volt anyway.
Figured they had to restrict it somehow. No way BMW would have allowed that loophole.
I’m a bit surprised it is that strict, though. Limiting the warranty to cars sold and currently registered in California makes sense, but the requirement that the vehicle is never registered out of state is a bit extreme. It would suck to void your warranty by moving across state lines for a few years.
It may not be worth it. I seem to recall that manufacturers tend to estimate warranty repair time…let’s call it “optimistically”. If you only get reimbursed two hours for a warranty job that realistically takes three you’re not going to seek out more of that type of work.
This sums up how buying a BMW changed DT.
I’ve gone soft!
Milk crates and spare tires not good enough for you now, Mr Fancy Pants (or shorts)?
That’s what she said….
Is there really anything better than a “that’s what she said” joke? No, there is not! COTD on COTD on COTD.
Wool tweed is the best upholstery. Change my mind.
Sit on it in wool pants in low humidity, watch the lightning bolts from your fingers as you hand gets near any metal parts of the car
This doesn’t sound like a negative. I’m positive that it sounds like a superpower.
Okay autopian poll. How soon before DT does an Grace and slaps plastic covers over his I3 interior?
DT recently gave me his jalapeños at dinner. “Something, something… delicate palette.”
Palate.
Palette = Bob Ross
Pallet = forklift Bubba