BMW i3 behavior when down to empty (aka limp or turtle mode)

[ChrisC]

A fellow BMW i3 owner recently did some tests of the car's behavior when run out of fuel, whether out of battery in a BEV i3 or also out of gas in a REx i3. He posted the results to our local (closed) i3 owners Facebook group, and I thought it would be worth posting here where everyone else can see it, and where forum search (and Google) will find it. Most people like to know how their car behaves when it runs out of fuel, so that they know how close they REALLY are to the end.

FYI, he has an SOC display in his i3, so he could see what the actual SOC was, and not have to rely on the range Guess-O-Meter (GOM).

First, here are his BEV i3 test results:

(As the battery ran down to empty) ... the GOM seems to always hit 1 mile remaining at 5-6% SoC. In this run-until-dead test it stayed at 1 mile until around 2%. So effectively, if you use the GoM to gauge range, it appears to provide you with around a 5% buffer.

Once you get under 3 or 4% SoC it starts taking away power blocks (the power gauge display changes to show reduced power available). At 2% battery and below acceleration is similar to a Leaf in turtle mode.

One thing to note - once it's down around 2% the heater doesn't work at all, even on defrost (the fan works but it blows only cold air).

(In response to question on how many miles he went once the GOM said 1 mile remaining) ... About 3.5 miles, but I was trying to get to 0 - that's 5% SoC, so you could stretch it to 5 miles if you had to. The GoM goes to 0 miles, then to "---", which is a big improvement over the Leaf which goes to the dashes at a much higher SoC (around 6% I believe). I still can't imagine running the battery down to this level with only the GoM to rely on.

My own summary interpretation of this is that for the BEV i3, you get 2-4 miles of reserve after GOM's predicted zero point, recognizing that the GOM sticks at 1 for a long time (longer than a mile). So if I see 5 miles on the GOM, I actually have 7-9 miles. The power gauge display shows less and less power available (less power "blocks") so it appears that you get LOTS of warning, if you know to watch that display.

Second, here are his REx i3 test results:

I picked up a REx loaner for a trip out of town this weekend, and got to test out the total range on the way home tonight. ~130 miles were at 70mph, and the car was carrying about as much cargo as an i3 can handle. No refueling or charging on the way.

(After running the battery down and going into REx mode, and then running the REx fuel tank empty) ... the car put itself in park as I pulled into my driveway, despite having at least 4-5% of the battery remaining (REx ran out of fuel about 2 miles from home). All of the ePower bars were gone, similar to the behavior of the BEV at 0% SoC. Turning the car off and back on restored all of the bars, and allowed me to drive the car out of the road and into the garage, so this seems to be an issue with the software on the REx.

So, for the REx i3, after the gas ran out it had another 2 miles of moderate speed driving, and then suddenly put itself into park despite having quite a bit of battery charge left. This buggy behavior (quite different from BEV model) seems like to be a bug that BMW would fix.

FYI, I did similar testing on the Chevy Volt back in 2012, and wrote all that up at the link below, if you want to compare. In short, after the Volt runs out of battery power AND runs out of gas, it continues to run on electric drive for a while. It has enough power for highway driving for about 3.5 miles, then a rapidly depleting amount of power, enough to pull off the road and that's it.
http://gm-volt.com/forum/showthread.php?12144-running-out-of-gas-aka-limp-mode

 

[mindmachine]

Thanks for the info, good to know. Strange about the Rex.

 

[jackt]

I suspect the rex keeps more reserve otherwise if you were to refuel it wouldn't have enough power to start the engine.

 

[jadnashuanh]

I suspect the rex keeps more reserve otherwise if you were to refuel it wouldn't have enough power to start the engine.

Keep in mind that neither version of the i3 ever uses all of the battery capacity while driving. I suppose, if you parked it at nearly zero indicated and left it for weeks, it could get into the reserve they maintain both at the bottom and the top of the battery, but I don't think it has anything to do with trying to start the REx. Between the upper and bottom boundary, with the 22Kw battery, you can only use something around 18.8Kw.

 

[i3atl]

A fellow BMW i3 owner recently did some tests of the car's behavior when run out of fuel, whether out of battery in a BEV i3 or also out of gas in a REx i3

Thanks for summarizing the results here, Chris!

I'll try and answer any questions you guys have about the behavior in these situations.

FWIW, regarding the REx situation, I don't think that it's intentionally leaving this additional buffer, because once the car was turned off and back on it returned to normal operation (just like the BEV) where all ePower bars were available. I actually turned the car off and on again one more time after regaining power, and still had all of the bars - so I think it shifting into park was just a software issue that BMW needs to correct, assuming this isn't desired behavior (not sure why it would be, given that it hadn't started reducing power at this point).

If I had continued driving after the restart, I expect that it would've followed the same BEV "turtle" procedure by gradually reducing available power and shutting down accessories.

 

[Ruffles]

What do you mean the REX shifted into Park? Like, while you were driving it did that??? Did it lock up the wheels? I must be missing something here because that sounds extremely dangerous and destructive.

 

[jadnashuanh]

Many batteries 'recover' a bit after you stop draining them, so at least momentarily after you've turned the car off, it could appear to have a greater battery SOC when turning it on again than when you turned it off. If so, it won't last long.

 

[i3atl]

Many batteries 'recover' a bit after you stop draining them, so at least momentarily after you've turned the car off, it could appear to have a greater battery SOC when turning it on again than when you turned it off. If so, it won't last long.

That's not what happened here - the SoC indicator did not move from the time I turned the car off to the time I turned it back on. I don't have the exact reading, since this car isn't coded, but it was around 3-4%. The car clearly decided that it needed to go into park, for whatever reason, even though it wasn't nearly at that point yet.

 

[i3atl]

What do you mean the REX shifted into Park? Like, while you were driving it did that??? Did it lock up the wheels? I must be missing something here because that sounds extremely dangerous and destructive.

You're not missing anything, except that this happened at very low speed (less than 1mph). Even at low speed, you're correct that it's not a pleasant result, but this functionality isn't unique to the i3. Opening the door with your seatbelt unlatched, while at very low speed, in drive or reverse, on any modern BMW will result in the same behavior. I'm not defending it, just pointing out that it's not unique to this model.

The problem is that it did this with plenty of energy remaining in the battery, where the BEV only gets to this "forced park" state at 0% SoC. The REx skipped the process of gradually removing the ePower bars (turtle mode) and just went straight to Park. It behaved properly once I restarted, though.

In reality this is very, very unlikely to be an issue in normal use, as a REx owner would have refilled the gas tank long before getting to this point.

 

[nowtta60]

You know if I had a USA spec i3 I would just get the REX coded to allow maintain state of charge mode. There's no way I would want the battery to run down first on long trips - say anything over 120 miles. Allowing you to run the gas down first means you've got time to find a place to fill up whilst running on pure EV mode.

You can also help save lives by not polluting city air at the end of your journey and potentially avoid being squashed by a big rig yourself when doing 45 in a 75!

You'll never need to worry about turtle mode and CARB can go stick their buts in the air ready for a beating with whatever sports bat you prefer.

My own commute is 120 miles (last job was 150 each way but only once a week) and there's no way I would have bought an i3 without the REX being able to keep the battery in a safe margin of charge.

All we need now is someone offering an after market 20-30 litre tank and the only serious flaw in the i3 would be history.

BEVx my arse! Its a series hybrid. Just admit it.

How about BMW give people a choice.. you can get the BEVx edition or a Series hybrid edition, and the difference in price would be the CARB credit they lose. Oh hang on... people would buy the BEVx and then get the coding done Which I'm sure they knew all along.

 

[imran]

I ran out of gas today on my 14 i3 Rex (not coded yet) and it was a really bad experience. Car stopped with 5% SOC remaining and would not move at all. Powering it on and off did not work either. I was on a steep winding ramp entering my work parking lot on 4th floor of the building. I had to stop the car just before the end of ramp but when I tried to drive it again, it put itself in park and would not move. I was less than 100 feet from charge station but on a steep ramp so car could not be pushed up there.
I had to call BMW roadside assistance who turned up after 90 minutes with a gallon of gas while I stood behind the car cautioning and guiding all the cars coming up the ramp. I will probably start carrying a small fuel can in the car in future.

 

[Bawareca]

Well, running out of gas with any car is a bad experience

 

[jadnashuanh]

I don't get the uproar some have on how the i3 works when you try to run it to empty! At least it warns you in advance, in more ways than an ICE does, if that even has a low-fuel warning light. Even the miles-to-empty indication on an ICE is often inaccurate. IT all depends on your driving style, speeds involved, the traffic, and the terrain (certainly, if you're 'coasting' downhill, the estimate would be wrong for either type!).

It is good to know in advance what to expect, and this thread helps, for those inclined to actually understand their vehicle. The dealers should educate people better, but then, how many people actually listen or take them up on it, or read the manual?

Could it be dangerous? IMHO, no more than running out of gas on an ICE. And, probably safer, as you typically can limp to the side of the road, which may be impossible if you run out on an ICE in stopped traffic.

 

[alohart]

I don't get the uproar some have on how the i3 works when you try to run it to empty!

If I had a REx whose engine ran out of gasoline and whose battery pack's charge level was 5% but it would not propel the car, I'd be very unhappy as well. There's no reason other than a software bug that would explain why an i3 at an actual 5% charge level would not drive normally after running out of gasoline.

It might be that the charge level is incorrect and is actually near 0%. Several i3 owners have reported this problem. The charge level measurement can be inaccurate, especially as battery pack's age, so it's risky to drive at a very low charge level.