Reason will only charge to 100%

BMW i3 Forum

Help Support BMW i3 Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
What use case(s) cause the pack to be so far out of balance that it can take a few months of charging to 100% to rectify?
I checked mine a while ago after only charging to 80% (via home charging) for quite a while and it was still perfectly balanced according to Electrified app. I don't use it daily however so I have assumed, rightly or wrongly, that balancing may occur at < 80% and the pack is balancing during the days it's not in use.

Does frequent fast charging to lets say only 60% and using the car daily or many times per day cause the pack to become out of balance over time? as not resting for long enough at a high(er) state of charge where balancing occurs more quickly?

Seems like charge level is only one part of maintaining a healthily balancedbattery pack
Agree with all that. My use case was the car was the daily driver almost every day for 5-15 miles only. Over 2 year. Once every month maybe a 150 miles trip with 3 rapid charges (60Ah) only ever topped up on free (gratis) 7kW in town during the daily driver session so never using any paid electricity such that the SOC was probably always between 25%-75%. Come the colder weather I get 35m on a full charge - so maybe 12 months of that use case resulted - at the time I had no measurement capability. But it recovered after charging at home when the ‘free’ civic 7kW started fee based charging.

It wasn’t until attending the meeting in BMW HQ Farnborough that the realisation that a top cells was actually reduced ‘after’ charging completed to aid pack balancing that a little light bulb came on. I had for months been stopping the charge as soon as our retail excursions were complete. Never getting to 100%.
 
That is a terrific explanation of the impact of multi cell systems and the need for balancing. The fact we only learned in 2016 that the BMW BMS was doing this was an eye opener - the how and exactly what the conditions are seems still in doubt but the early adopters in our group achieved much better ranges than I did because they all charged to 100% at home whereas I was in Yorkshire and too stingy. I only ever charged off solar or the free local authority posts ( now long gone). I didn’t need 100% for range except on those DC charged forays where we stopped well before 100 as it took too long past 80% in the 60Ah. Splash and dash. So most other folks were using the fine ABC algorithm that prevented anyone discovering this little ‘out of balnce’ symptom so ably described by @alohart with the 2000 Honda Insight.
 
I also found this to add to the debate on degradation. Has been in my archive but someone was doing these tests in the USA. AVT-INL? Anyone? Static capacity test? It looks like the format from David’s book but I haven’t been back to check and why there are only 4 cars?
 

Attachments

  • IMG_3437.jpeg
    IMG_3437.jpeg
    65 KB
Different car, but my Kangoo EV has 50 bms cell-modules, each equipped with a 1 Watt dissipating resistor when balancing. It takes at least a few hours to balance the pack. Of course not every resistor need to dissipate electrons in the form of heat, but per 200Ah cell, it burns about 4Wh of energy per balancing cycle. Just a rough estimate of course, but if you have only one resistor balancing per full charge, it can take a while, like in a i3.
 
but if you have only one resistor balancing per full charge, it can take a while, like in a i3.
But in an i3, we have one resistor per module - hence 8 modules can balance simultaneously (assuming they all have cells that are far enough to ~need~ balancing, of course)?
 
I also found this to add to the debate on degradation. Has been in my archive but someone was doing these tests in the USA. AVT-INL? Anyone? Static capacity test? It looks like the format from David’s book but I haven’t been back to check and why there are only 4 cars?
The Advanced Vehicle Team at the Idaho National Laboratory (U.S.) tested all sorts of BEV's and HEV's. I was disappointed by how rapidly the usable battery pack capacities of all 4 i3 BEV's and 4 i3 REx's decreased which was my experience as well. This is the i3 section:

BMW i3 (2014) - Electric
Battery Pack Laboratory Testing Results
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
DC Fast Charging at Different Temperatures
Infographic
Report
Maintenance and Repair History
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Mileage Accumulation and Fuel Economy
On-Road Usage and Performance Summary
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Operating Costs
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Steady State Vehicle Charging Fact Sheet
Vehicle Specifications and Testing Results
BMW i3 REX (2014) - Electric
Battery Pack Laboratory Testing Results
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Maintenance and Repair History
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Mileage Accumulation and Fuel Economy
On-Road Usage and Performance Summary
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Operating Costs
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Vehicle Specifications and Testing Results
 
I believe that we didn’t have that as a specific ‘thing’ - although the ‘eflow’ app by the bimmerlink folks does give info that I think I have posted. Gives max and min voltages and average so we can see how far out it is but not the actual 96 voltages. As far as I know.
I beta-tested early eFlow versions. At least for several versions, the voltages of each battery cell could be displayed in a bar chart of voltage vs battery cell number. I believe touching the min-max voltage area resulted in this bar chart being displayed. I hope that this is still possible.
 
I beta-tested early eFlow versions. At least for several versions, the voltages of each battery cell could be displayed in a bar chart of voltage vs battery cell number. I believe touching the min-max voltage area resulted in this bar chart being displayed. I hope that this is still possible although I doubt that knowing which cell's voltage was low is of much value.
 
But in an i3, we have one resistor per module - hence 8 modules can balance simultaneously (assuming they all have cells that are far enough to ~need~ balancing, of course)?
So we have 12 cells per module, every cell has to be balanced somehow, so there should be a mechanism that assigns the module resistor to a cell to be balanced, if there is only one resistor per module. And then the modules themselves should be balanced against each other also. I don't know the details, however. Big mystery to me lol
 
So we have 12 cells per module, every cell has to be balanced somehow, so there should be a mechanism that assigns the module resistor to a cell to be balanced, if there is only one resistor per module. And then the modules themselves should be balanced against each other also. I don't know the details, however. Big mystery to me lol
Yes, so it's essentially a one-in-twelve operation. In terms of the details, we're all guessing, of course. AFAIK the exact mechanism has never been published by BMW so that is inevitable.

As far as I can tell, they don't balance at the module level. I'm only inferring that from the fact that there is no documentation that refers to an overall module-balancing resistor. My guess is that the "averaging" effect of balancing each cell against its 11 neighbours is good enough to keep the modules well-balanced compared to one another - but that is of course just a guess...
 
I beta-tested early eFlow versions. At least for several versions, the voltages of each battery cell could be displayed in a bar chart of voltage vs battery cell number. I believe touching the min-max voltage area resulted in this bar chart being displayed. I hope that this is still possible.
I’ll try that.
 
The Advanced Vehicle Team at the Idaho National Laboratory (U.S.) tested all sorts of BEV's and HEV's. I was disappointed by how rapidly the usable battery pack capacities of all 4 i3 BEV's and 4 i3 REx's decreased which was my experience as well. This is the i3 section:

BMW i3 (2014) - Electric
Battery Pack Laboratory Testing Results
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
DC Fast Charging at Different Temperatures
Infographic
Report
Maintenance and Repair History
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Mileage Accumulation and Fuel Economy
On-Road Usage and Performance Summary
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Operating Costs
Vehicle #5486
Vehicle #5626
Vehicle #5655
Vehicle #5658
Steady State Vehicle Charging Fact Sheet
Vehicle Specifications and Testing Results
BMW i3 REX (2014) - Electric
Battery Pack Laboratory Testing Results
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Maintenance and Repair History
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Mileage Accumulation and Fuel Economy
On-Road Usage and Performance Summary
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Operating Costs
Vehicle #3436
Vehicle #4162
Vehicle #4410
Vehicle #4557
Vehicle Specifications and Testing Results
That’s terrific. I think that is where David was drawing the data from for his book - He referenced all his sources. Wonder where those vehicles are now.
 
Pack imbalance is probably correlated better with the condition of the battery cells. If the internal resistance of a cell is higher than the others, more of the charging/discharging current passing through the cell would be converted to heat rather than to energy. The result would be that this cell would have a lower charge level when the pack is fully charged.

Cell capacities are never perfectly matched, so those cells with lower capacities won't have the same charge levels as those with higher capacities.

The temperatures of the cells will not always be the same, maybe due to their positions in the pack and how well cooling effects each cell. This temperature difference would affect the cell performance which would gradually result in charge level differences.

All battery packs with cells connected in series need to have the cells' voltages equalized, usually at a high charge level ("top balancing"). Our 2000 Honda Insight's BMS did not include cell charge level balancing, so the charge levels of its 120 battery cells gradually diverged over time. As a result, the usable capacity gradually decreased as the BMS prevented the cells at the highest charge levels from overcharging even though cells at lower charge levels could have accepted more energy, and when a cell's voltage decreased to the lower limit allowed by the BMS, other cells at higher charge levels were unable to discharge farther. When the usable capacity decreased to a certain level, a diagnostic trouble code (DTC) was stored and a battery pack trouble light illuminated. Honda's only fix was to replace the battery pack even though the battery cells weren't damaged but only needed their charge level's equalized. Insight owners believed that the battery packs that Honda was installing under warranty were packs rebuilt with used cells that had all been fully charged and thus were in balance. The battery pack in our Insight was replaced twice under the 8-year battery pack warranty.

A couple of enterprising Insight owners designed and build battery pack balancers that took advantage of the fact that NiMH battery cells aren't damaged significantly when occasionally overcharged. They merely convert the energy to heat, so the battery pack cooling fan was activated to prevent overheating during balancing. So charging continued until all cells were fully charged and thus their charge levels were balanced. I bought one of these cell balancers and began using it shortly after Honda installed my 3rd battery pack just after my battery pack warranty expired. This battery pack was still performing well 9 years later when we sold our Insight along with the cell balancer.

Fortunately, the i3 implements a much more sophisticated passive cell balancing protocol that seems to work very well.

I wish that BMW had provided some guidance in this respect. I think the good news is that cell balance seems to be maintained well regardless of how the battery pack is treated. Gonville's experience with a badly unbalanced battery pack seems to be unusual, or maybe he's one of the few i3 owners who would have realized that cell imbalance was reducing the usable battery pack capacity.
@alohart - may I borrow and attribute this explanation for the i3 group in the UK? I have been writing about ‘passive top end cell balancing’ until my heart bleeds but your explanation of the Honda process was very well written and should enlighten folks. Maybe the fact that 80% isn’t the golden solution to longevity but that charging to full occasionally could be of more value in retaining capacity. Whilst checking with the available apps the actual cell voltage variance.
 
Back
Top