Latest E400 BMS update appears to have reduced usable capacity.

[IanL]

The recall was done on my 2018 MY recently. Prior to the update, my SOH was about 101%. I calculate it using displayed SOC percentage before and after charge, mileage between charges, and kWh/mile between charges. The results vary a little, because of the poor granularity of the data, but are reasonably consistent. The resulting capacity in kWh is then expressed as a percentage of 30 kWh (the usable capacity) to give the SOH. I also record GOM before and after charge, and calculate "GOM pessimism".

Since the update, the SOH has dropped to about 92%,. The GOM pessimism has reduced correspondingly. In both cases, the change is marked, not a gradual movement. The explanation I have arrived at is that the relationship between the real SOC in Ah and the displayed SOC in percent has been altered, and the GOM algorithm correspondingly altered so that the displayed range appears much as before (which it does). The data shows that the usable capacity has been reduced from 30 kWh to about 27.5 kWh, as the BMS will not charge over the 100% displayed value.

Another item of confirmatory evidence is that the 5hr 20min charge (with my wall box set to 16A) I usually employ to get approximately 50% SOC increase, now gives 58-62%. Some variation can be expected, mainly due to ambient temperature, but the noticeable step change is not linked to a marked shift in the weather. I do not believe the box is providing more Ah, so the change must arise because of an alteration of displayed %age charge.

I think KIA may have done this to extend the battery life, as the 70% SOH replacement value would appear to have been altered from 21 kWh to about 19.25 kWh. It may also have benefit in reducing the probability of dangerous overheating, which is the stated purpose of the recall.

I would be interested in knowing if anyone can confirm my observations. One way would be to compare cell voltage values at 100% SOC before and after the update.

 

[JejuSoul]

The recent recall updated the BMS software. It is quite likely that the algorithm used by the BMS has changed. The PIDs we use to query the BMS also sometimes change.

But the GOM software is in the Navi computer. The trip data is stored here. The software for this has not been changed. The change must be that the BMS is sending different data for the SOC at the start and the end of a trip.

The SOH that should be used is from the original usable capacity of 30kWh. This is what the battery warranty applies to. It would not be okay for them to lower the 70% SOH replacement value by later artificially reducing the usable capacity.

 

[IanL]

... It would not be okay for them to lower the 70% SOH replacement value by later artificially reducing the usable capacity.

I quite agree. I will watch SoulSpy for when the reported SOH drops below 100% and then check my value for capacity. As Spy uses the SOH value computed by the BMS (which is presumably also used by KIA for warranty replacement), that should reveal whether they are playing it straight.

Do you perhaps have a record of cell voltage when fully charged before the recall was applied? Unfortunately, I do not.

 

[JejuSoul]

Previously I have written this :- Analysis of a replacement battery.

Also of note - a MY2018 30kWh Soul EV has a final resting voltage at 100% SOC of 410.0V
This has 100 cells, each charged to 4.10V
My car has 96 cells, each charged to 4.14V

I assume this means there is a much larger buffer on the MY2018 30kWh Soul EV than a MY2015 replacement battery 27kWh Soul EV.

The data came from 3 cars.
My blue 27kWh 2015 Soul EV. Original BMS. Data read 4 years ago. SOH = 96%
My white 27kWh 2015 Soul EV after replacement battery. 2nd BMS. Data read 4 years ago.
Norwegian 30kWh Soul EV. Original BMS. Data read 6 years ago.

There have been plenty of BMS updates in the intervening years. The latest being the 'Recall BMS'
that most of our cars now have. The BMS software is not the same. There is no way I can keep track of all the changes. Hence I no longer track BMS data for cars other than my own. Nor do I recommend trusting any app that gives BMS data, other than the one that Kia itself uses.

What I do recommend if you intend keeping the car for the long term is to track the range data over many seasons. This would give a real trendline of the car's usable capacity to compare the BMS data against.

 

[IanL]

Thanks, the data for the Norwegian Soul is just what I need. I will report when I have next done a 100% charge.

 

[IanL]

Well, I have done a 100% charge , and the battery voltage is 413.6, with all cells reading 4.12 volts. This was after 12 hr from full charge end. The reported SOH, stated as read from the BMS, is 100%. All data from SOULEV Spy.

This does not support my theory. I will have to search for another explanation.

 

[chilibt]

Battery update (august 2024 version) removes range in 2018 Soul EV.

After the first Battery software update in February removed 20% of my range an new update was given by KIA as a "Voluntary recall" dated August 2024.
Now it has been 17 days and here is the fact for my car:
The new update maximizes the range to only 210 KM. My car was getting a summer range of 240 to 255 Km, now it is limited to a Maximum of 210Km. No highway driving, Max speed (no AC) 80 Km h.

My EV spy software showed in early August a max charge as 4.18v per cell prior to the newest update (I do not have the data from Feb, prior to the first update) Now it is limited to 4.10 volts, and never above! (both sets of data was at a cell temp of 25C-27c)

As a result of the first update(Feb 2024) I lost 20% range (max range 196). Now the new August update gave me a loss of 16% from last year. (242km avg down to 210 km). To add to the problem, the range data is terribly inaccurate. My 210 range yesterday could not get me 180Km, without having to find a level 3 charger to get home. (80 kmh max , no AC used).

Another change is the efficiency rating (scale of 1 to 8). I was consistently getting 7 or 8 for 5 years. Now I see 4 to 6 all the time and only 8 for a few minutes a day. it seems the range and efficiency calculations are unstable, often I see one range on shut off, and then a few minutes later start up and it is 5 Km range different. Once I saw the range drop by 10 km instantly as I drove.

I believe after 5.5 years of ownership of this car, having a OEM company modify a car to degrade its range is against consumer protection principles. I am thinking that KIA is degrading batteries to reduce its risk of replacing batteries. This reduces the range that was sold to consumers below to what was its "stated" range. Does anyone else have the same experience and concerns? (FYI we loved the car UNTIL the updates, now it is tainted to us, all because of software).

Also the original recall was for "fire" risk, and the harness was to be inspected. Does anyone know why a degrading of the battery cell voltage was supposed to reduce the risk? I cannot find the reasoning/explanation on the BMS changes and what they were supposed to do to reduce the risk.

 

[IanL]

Thanks for your post, which confirms my findings. I believe my loss of range is of the order of 14%, based on my capacity calculations, but, in my case, there is no indication of this on the GoM, as it has suddenly become more optimistic, thus showing no decrease of predicted range.

 

[IanL]

I would not be surprised if, in view of your odd GoM behaviour, you are offered a further update. The update I received is clearly different in that regard, and I suspect they have been experimenting, and have now (in my case) got their "hack" to work the way they intended.

 

[SoulEVSpy EVRanger]

When BMS firmware is updated, the SOH estimate in the BMS is wiped. The GoM is based on SOC multiplied by SOH, so your GoM will be based on assumed 100% SOH, until the BMS has made a new estimate of SOH, after a number of full discharge / recharge cycles.

 

[IanL]

Thanks. I will run down to 18% and do a 100% charge and see what transpires.

I can see from my records, by comparing (for each recharge) GoM added to %charge added, that a substantial change occurred since the update and is persisting. I attributed this to apparent re-scaling of charge percentage, but it could also be affected by different behaviour of the GoM. However, against that idea, my SoH is still at 100%, so the resetting of the GoM algorithm should have had no effect.

 

[JejuSoul]

The recent recall updated the BMS software. It is highly likely that the algorithm used by the BMS for your car a 2018 has changed. The algorithm used by the BMS for my car a 2015 has definitely changed.

The 27kWh cars with replacement battery got new BMS hardware and a new wiring harness.
The 2015 cars got 2018 hardware and 2018 software.
One noticeable effect is that charging to 80% now stops at 83%.
Probably the algorithm is stopping at where 80% would be if the car had a 30kWh battery.

The 30kWh cars with or without replacement battery just got the new BMS Sofware.

Resetting the BMS software automatically resets the values stored in the BMS.
This is not an algorithm change.

The GOM software is in the Navi computer. The trip data is stored here. The software for this has not been changed.

I changed my opinion from quite likely to highly likely for the algorithm change on the 2018 because we now have two datapoints for such a change. There is no way for us to know for sure because Kia will never tell us.

We have no knowledge of how any of the BMS Sofware changes mentioned in this thread relate to preventing fires.

 

[mariano3113]

The recent recall updated the BMS software. It is highly likely that the algorithm used by the BMS for your car a 2018 has changed. The algorithm used by the BMS for my car a 2015 has definitely changed.

The 27kWh cars with replacement battery got new BMS hardware and a new wiring harness.
The 2015 cars got 2018 hardware and 2018 software.
One noticeable effect is that charging to 80% now stops at 83%.
Probably the algorithm is stopping at where 80% would be if the car had a 30kWh battery.

The 30kWh cars with or without replacement battery just got the new BMS Sofware.

Resetting the BMS software automatically resets the values stored in the BMS.
This is not an algorithm change.

The GOM software is in the Navi computer. The trip data is stored here. The software for this has not been changed.

I changed my opinion from quite likely to highly likely for the algorithm change on the 2018 because we now have two datapoints for such a change. There is no way for us to know for sure because Kia will never tell us.

We have no knowledge of how any of the BMS Sofware changes mentioned in this thread relate to preventing fires.

I think my 2015 Soul EV got a 30kWh battery as post recall BMS and wiring harness it ended up still needing a battery replacement.

Current battery has had the most range and DC charging to 80% is till 80% not 83%.


However my car has been back at the dealer since Aug 26th because it has rolled out of the driveway twice since new battery & new parking brake recall.


Luckily it was not plugged in & charging, when it rolled into street (cul-de-sac).


Currently have an EV loaner as I await my Soul EV to come back. (Been gone most of this year Dec - May and then June to August and now again August until xxxx.)

Have had my Soul EV for less than 15 days this year.

 

[IanL]

Thanks. I will run down to 18% and do a 100% charge and see what transpires.

I can see from my records, by comparing (for each recharge) GoM added to %charge added, that a substantial change occurred since the update and is persisting. I attributed this to apparent re-scaling of charge percentage, but it could also be affected by different behaviour of the GoM. However, against that idea, my SoH is still at 100%, so the resetting of the GoM algorithm should have had no effect.

I have run down to 14% and recharged to 100%. Spy reports most cells at 4.12V, with a small block at 4.10V. My calculated SoH is 88.5% (Spy still reports 100%), and GoM is consistent with the eight previous records after the BMS update and the m/kWh displayed. That suggests that GoM recalibration had no effect, and the observed behaviour is consistent with fudged SoC values passed to the NAV computer to hide the reduction in maximum SoC.

More food for thought:

Spy shows battery SOC_display_pct at 100
Battery SOC_pct at 95

Car Scanner shows SoC display % at 100
SoC BMS % at 95
SoH 1% at 100
SoH 2% at 60

Car Scanner has not previously offered any SoH parameters, suggesting that the app's profile for the SoulEV has been updated. No idea what SoH 2 is (maybe the auxiliary battery?).

The difference between the two SoC values supports the altered maximum SoC hypothesis.

 

[chilibt]

A new problem occurred. Sept 3 BMS update. This week after charging and a short trip, I started up and the range went to 223, which is 20 Km more than the start up max of 210 just minutes earlier. My range max before the first BMS update was 255, averaging 242 in summer/fall conditions.
This week, at the same time it showed 226, the range estimate was very accurate, which is not the case every time I start since the BMS update. I took a evSpy reading and the battery voltage was 4.14

Also this week my re-gen braking failed to work. I was getting close to an accident as the behaviour of the car made me think my brakes failed. I pulled over later and shut down and restarted and the re-gen was back. So watch out this is another software issue.

 

[IanL]

I had that braking failure once, but before the update. Same "cure". Switched off and on again and it has never returned. When in doubt, reboot the computer. I don't think the update is relevant in this case.

 

[Nigel]

Had my second BMS software update on my 2016 Soul last week with some trepidation due to postings here indicating range loss and other problems. In my case the car went to the dealer with 40km range left and after driving home it was down to 20km. Put the car on charge and took it to 100% and the GOM indicated a range of 226km on a 27kWh battery pack. You can imagine that I thought this was an error so I drove the car down to 18km range and had driven just over 200km so it looks like a real range increase. Charged the car to 81% and it had an indicated range of 179km. Going to run it down to 20km or thereabouts and see if this is real as well. Don't understand where the extra range is coming from unless Kia have unlocked some reserve they might have in the 27kWh E400 battery.

 

[JejuSoul]

Had my second BMS software update on my 2016 Soul last week with some trepidation due to postings here indicating range loss and other problems. In my case the car went to the dealer with 40km range left and after driving home it was down to 20km. Put the car on charge and took it to 100% and the GOM indicated a range of 226km on a 27kWh battery pack. You can imagine that I thought this was an error so I drove the car down to 18km range and had driven just over 200km so it looks like a real range increase. Charged the car to 81% and it had an indicated range of 179km. Going to run it down to 20km or thereabouts and see if this is real as well. Don't understand where the extra range is coming from unless Kia have unlocked some reserve they might have in the 27kWh E400 battery.

It is now over 4 1/2 years since I had the battery replaced on my white 27kWh 2015 Soul EV.
Am still getting range like yours.

I think my 2015 Soul EV got a 30kWh battery as post recall BMS and wiring harness it ended up still needing a battery replacement.
...
Current battery has had the most range and DC charging to 80% is till 80% not 83%.
...

The replacement battery for a 2015 Soul contains 96 E400 cell-pairs.
This is not a 30kWh battery pack. That has 100 E400 cell-pairs.
But as you can see from the range figure above it acts like a 30kWh battery pack.
It is AC charging when set by the timer that changed not DC. It finishes at 83% not 80%.

 

[mariano3113]

It is now over 4 1/2 years since I had the battery replaced on my white 27kWh 2015 Soul EV.
Am still getting range like yours.
View attachment 209



The replacement battery for a 2015 Soul contains 96 E400 cell-pairs.
This is not a 30kWh battery pack. That has 100 E400 cell-pairs.
But as you can see from the range figure above it acts like a 30kWh battery pack.
It is AC charging when set by the timer that changed not DC. It finishes at 83% not 80%.

That is nice to know.
I assumed the 37510-E4203R battery might be 2015-2019 replacement battery pack instead of only replacing batteries from 2015-2017 models.

Then again I think there was only a 1.3kWh total pack size deviation...so 27kWh pack was closer to 30.5kWh vs 31.8kWh for the 30kWh 18'-19' Soul EVs.

 

[JejuSoul]

That is nice to know.
I assumed the 37510-E4203R battery might be 2015-2019 replacement battery pack instead of only replacing batteries from 2015-2017 models.
...

The pack 37510-E4203R has 96 cells and is for the 2015-2017 models.
The pack 37510- E4302R has 100 cells and is for the 2018-2019 models.

Given that the fire safety recall adds the 2018-2019 BMS hardware and wiring harness to the 2015-2017 cars I'm not sure why they bother keeping the two versions.