Any data on heat vs battery capacity?

bwilson4web · Jul 18, 2017

Hi,

With the summer heat, the BMW i3 Facebook is seeing more posts of low capacity batteries in high temperature environments. But I'm wary of ad hoc, one-of, postings because they can reflect: (1) abused by excessive "mi/kWh"; (2) incomplete temperature records, or; (3) it broke.

With warmer temperatures, my BMW i3-REx battery seems to have gained longer range, 80-95 miles. So I'm wondering if there have been any studies of i3 capacity vs local temperature?

One thought is to observe the indicated range at dawn after cold-soak all night. Then go out at sun set after an afternoon heat soak to see if the indicated range has changed. I can use local climate data to get an idea of the heat soak. But I'm not sure this would give any useful information.

Does the BMW i3 turn on the battery air conditioner by itself in high heat?

If preconditioning departure time is set, how early can the car begin the cool down?

Sorry to be rambling but trying to figure out the BMW i3 battery thermal management seems somewhat obscure. Perhaps there is a study or article out there that might insights to what others have seen?

Thanks,
Bob Wilson

alohart · Jul 18, 2017

bwilson4web said:
Does the BMW i3 turn on the battery air conditioner by itself in high heat?

It does when the battery pack is being charged or the car is in Ready mode, but not when it is parked in the heat not charging or not in the Ready mode.

As can be shown, turning on the cabin A/C reduces the GOM range estimate, so I would be surprised if a high battery pack temperature would not reduce the GOM range estimate. This range estimate reduction would be mostly because the A/C compressor would need to be running to cool the battery pack, not for electrochemical reasons.

theothertom · Jul 18, 2017

bwilson4web said:
With warmer temperatures, my BMW i3-REx battery seems to have gained longer range, 80-95 miles. So I'm wondering if there have been any studies of i3 capacity vs local temperature?

Just another data point....my last charge to 100% was about a week ago and the GOM read 149 miles. Previously it read in the 120-130 range. I have noticed the miles/KWhr have increased from 3.9 to 4.0. I charge my car in my basement which is relatively cool.
I drove the car for several days around town, not on a long trip, until the REx came on after 112.6 miles.

There is a guy who posted here about a book he had written on the i3, which contains a lot of info on the battery. Here's a link to the post:
http://www.mybmwi3.com/forum/viewtopic.php?f=2&t=4424

jadnashuanh · Jul 18, 2017

When you set a departure time, the battery conditioning can start as much as 3.5-hours prior to your announced departure. The cabin conditioning, should you check that box, typically doesn't start until 20-25-minutes prior to the departure time. If it needs to heat things up, it uses a resistance heater that is rated at 1Kw. I don't know the actual cooling capacity of the battery cooling circuit nor the cabin cooling capacity.

bwilson4web · Jul 20, 2017

Hi,
A couple of follow-up questions and sharing data:

jadnashuanh said:
When you set a departure time, the battery conditioning can start as much as 3.5-hours prior to your announced departure. The cabin conditioning, should you check that box, typically doesn't start until 20-25-minutes prior to the departure time. If it needs to heat things up, it uses a resistance heater that is rated at 1Kw. I don't know the actual cooling capacity of the battery cooling circuit nor the cabin cooling capacity.

Earlier this week, the car was parked in an alley in the sun, +90F, with the windows up. While I had lunch, the car charged as expected and I noticed the taper down. So I paid my bill, checked the charge and noticed it had just dropped to 0 amps.

I had to do some shopping at the corner grocery store and remember selecting the first option, "Activate Now" to cool the cabin. When I unplugged the car, the 'blue' light was blinking and drove home. It was later when I noticed the 'cool down' profile.

The chart shows the cool down ramped up to ~4.5 kW and then down to ~1.5 kW which it held until I unplugged. The high initial rate suggests it might have been trying to cool the battery but I'm not sure.
Pure speculation on my part but I wonder if "Activate Now" may follow a cool-down protocol:

Check cabin to start either cooling or heating, in this case, cooling.
Check battery temperature, if too warm, start cooling.
As battery temperature comes down, reduce cooling.
Continue cabin cooling as needed based on temperature measurements.

Summer in Alabama, I can repeat the lunch test only next time use the "Set Departure Time" instead. If the cool-down profile changes, we'll have data to share. Being retired, I can set my own schedule but I'm open to suggestions to improve the test.

Going from home to Propst:

6.7 miles - leave at 10:40 AM, ~1.5 kWh @4.5 mi/kWh
Plug-in at 11:00 AM - when my favorite restaurant opens
Lunch - wait for charge current to drop to 0 A, should take ~20-30 minutes
Wait 15 minutes
Schedule departure time one hour out (to see if it starts right away or waits)
Hang until the EVSE shows 0A or 2:00 PM, last call for lunch (bring a good book)

Bob Wilson

bwilson4web · Jul 21, 2017

Hi,

So I ran the test today and got the following:

~11:10 AM - plugged in the car after driving 6.5 miles in 16 minutes. At 88F (31C) on the driveway, it was 92F (33C) when I parked in the alley to plug-in. I then put in our prescription refills and went to lunch.
~11:40 AM - I noticed charging had stopped so I scheduled a departure time of 12:45 PM
~11:45 AM - within a minute of sending the departure time, there was a five minute charging session, 1.72-1.42 kW. I've seen these short, charge bursts in the past so I suspect is it not related to getting the departure time message.
~12:18 PM - charging bursts up to a peak of 5.8 kW and in 10 minutes down to 1.8 kW
~12:45 PM - down to 1.4 kW while I was getting our prescriptions
~12:50 PM - down to 1.3 kW when I entered the car, I could hear the A/C working and the interior was nice.

I loaded the chart into a spreadsheet and digitized the curves relative to 00:00:00 (hh:mm:ss):
Time,kW
00:00:00,0.00
00:02:40,0.00
00:03:15,6.48
00:12:30,6.40
00:16:30,3.50
00:21:00,1.85
00:25:00,1.40
00:30:00,0.85
00:36:15,0.60
00:36:30,0.00
00:45:00,0.00
00:45:25,1.72
00:50:00,1.42
00:50:15,0.00
01:05:00,0.00
01:18:00,0.00
01:18:20,5.80
01:21:00,3.10
01:27:00,2.30
01:27:15,1.80
01:38:00,1.60
01:40:00,1.55
01:45:00,1.40
01:50:00,1.30
01:50:15,0.00
02:00:00,0.00

Initial Thoughts

"Activate Now" - on Wednesday, there was not enough time to let the battery reach a state where the second charge pulse could occur independently. So the triangular shaped charge probably is a combination of battery thermal management followed by a decay to the 1.4-1.8 kW range when the cabin cooling becomes a steady state load.

"Set departure time" - by having a one hour delay, the initial battery charge and probably thermal delayed, final charge, occurs letting the car achieve a steady state while parked. Then based upon the set time, about 20 minutes before departure, there is jump to 5.8 kW that is probably a battery cooling, initial topping charge, and possibly the cabin (not observed.) After 10 minutes there is step down to 1.8 kW that at the scheduled departure time is 1.4 kW and five minutes later, 1.3 kW.

I suspect the longer preconditioning lead times, ~3 hrs, occur in cold weather when the battery has to be warmed up. A cold-soaked battery probably needs the longer time to reach an optimum operating temperature.

In contrast, these hot weather tests did not include a heat-soak. I had to drive to the recording EVSE, ~15 minutes giving an initial, operating cool-down. I started in the shade with external temperatures of ~90F (32C) and then parking in a sun lit, alley 100F (37C). There was not enough time to properly heat the battery to ambient temperature.

That is what the data suggests to me given we're limited to just the EVSE charging data.

Bob Wilson