MG4 51kWh LFP Battery Health

[AlunJ]

Had my SR back from its 2 year service. Last year at 15k it had 95% SoH… this year at 26k… 98%. Go figure

 

[fnegroni]

Had my SR back from its 2 year service. Last year at 15k it had 95% SoH… this year at 26k… 98%. Go figure

You simply charged it from low Soc to 100% and when they report back the Soc they might have an adjustment on top of what’s returned by the OBD2.
Do you mind checking?

 

[Archev]

Had my SR back from its 2 year service. Last year at 15k it had 95% SoH… this year at 26k… 98%. Go figure

Package that and sell it boy

 

[AlunJ]

You simply charged it from low Soc to 100% and when they report back the Soc they might have an adjustment on top of what’s returned by the OBD2.
Do you mind checking?

I don’t have an obd2 dongle sorry. You could well be on the money though. I used to charge it every night after work but I’ve moved to a slightly closer depot so do sometimes let it drop lower now.

 

[siteguru]

However ... I recently did a 4% to 100% and balance charge, with no difference whatsoever to the SOH reported in Car Scanner.

 

[Bam Bam]

However ... I recently did a 4% to 100% and balance charge, with no difference whatsoever to the SOH reported in Car Scanner.

Is it something you do regularly, though? In which case it might not have made such a difference.

I rarely do a balance, unfortunately, because we rarely go on a trip long enough to get down below 10% and need well over 20% for wife to get to work and back so it usually has to get charged when on 18% or something at the lowest.

 

[siteguru]

Maybe 2-3 times a year I do it, if the opportunity arises. At 1st service my SOH was 95%. At 2nd service it was 93%. After getting an ODB2 dongle and Car Scanner I saw 92.8%, and it slowly dropping over time. (It was 92.65% when I last checked a few weeks ago).

So I simply don't believe that the LFP pack will recover 3% of SOH in a year, regardless of what charging regime you've followed. Especially since @AlunJ doesn't appear to have done any such calibration charge.

 

[fnegroni]

However ... I recently did a 4% to 100% and balance charge, with no difference whatsoever to the SOH reported in Car Scanner.

One might not be sufficient, plus surely if you already did it regularly there might be no change in your case.

I charge from <10% to 100% once a month.

However ... I recently did a 4% to 100% and balance charge, with no difference whatsoever to the SOH reported in Car Scanner.

Is your CarScanner config up to date btw?
If you share your settings here I can tell you if it’s not up to date.

 

[siteguru]

Please see my reply above. Plus pretty much all my AC charging (which accounts for >98% of all charging) is from 30% or lower SoC to 100% and balance. So I reckon I'm giving my pack the best chance for longevity vs convenience.

(Was typing as your next reply came in). I'm on the phone just now but will check ASAP. What specifically are you wanting to see? And do I need to connect to the car to get access to such details, or will the info be there in the app anyway?

 

[fnegroni]

Please see my reply above. Plus pretty much all my AC charging (which accounts for >98% of all charging) is from 30% or lower SoC to 100% and balance. So I reckon I'm giving my pack the best chance for longevity vs convenience.

As I said, you might really have a pack that is at 92%.

(Was typing as your next reply came in). I'm on the phone just now but will check ASAP. What specifically are you wanting to see? And do I need to connect to the car to get access to such details, or will the info be there in the app anyway?

the connection config screen had to be manually set to Send FF to avoid data being misread or not read at all before timeout.
The developer recently updated the parameters and also some of the sensors.

 

[siteguru]

I'll check when I can. I do know the last time I used it the app said something about an update (which I allowed).

 

[AlunJ]

So I simply don't believe that the LFP pack will recover 3% of SOH in a year, regardless of what charging regime you've followed. Especially since @AlunJ doesn't appear to have done any such calibration charge.

Yeah I don’t buy it either. More likely mistaken the reading and 93% would make more sense in my case.

 

[siteguru]

the connection config screen had to be manually set to Send FF to avoid data being misread or not read at all before timeout.
The developer recently updated the parameters and also some of the sensors.

Where do I find the Send FF item please?

In the Advanced Settings I can see the following for Custom initialization string:
ATZ
ATE0
ATH1
ATSP0
ATS0
ATM0
ATAT1
ATSTFF

Is it that last one?

 

[fnegroni]

Yep that’s correct now.

 

[siteguru]

In which case the last SOH reading I looked at was based on those settings.

 

[AlunJ]

Yeah I don’t buy it either. More likely mistaken the reading and 93% would make more sense in my case.

Bought a dongle and car scanner reports 92.99% so as suspected closer to 93 than 98%

 

[Tig170]

I find the most reliable way is to read the total energy consumption from last charge on the display, I always charge to 100% so when its at a simple marker point e.g. 50% battery left, if my total consumtion is 24kWh (it was last time I checked), then I know that the car is using around 48kWh from its unbuffered? 51kWh max capacity - for comparison car scanner says its 93% SoH (but its just a linear trend with mileage or time, as ive monitored it since 5000 miles ) - and the dealer said it was 100% ( car done 30k miles now )- this is a useful watch:
I note that new ones have a capacity of 49kWh, I wonder if they have just changed the software buffer rather than made any physical changes (Kia do this apparently)

 

[Coulomb]

I know that the car is using around 48kWh from its unbuffered? 51kWh max capacity...

Yes, 51kWh is nominal, unbuffered capacity.

Though according to ev-database, the usable capacity is 50.8kWh, only 0.4% lower than nominal. For comparison, my 77kWh ER has a usable capacity of 74.4kWh, making the buffer 3.4%, some 8.5x larger.

 

[king31]

I share with you my research on the gross capacity of the MG4 SE 51 kWh LFP. The current SOH is only an estimate, as the car continues to charge at full capacity. Here are the details regarding the usable and gross capacity.

MG advertises 50.80 kWh of usable energy, but in reality, it is much more. According to the datasheets, the usable capacity is 156 Ah for a weight of 398.7 kg and 104 cells in series. If we calculate with a nominal voltage of 3.2 V per cell, this gives 51.92 kWh "usable" (I emphasize this term).

However, the CATL cells used have a higher nominal voltage of 3.35V at 99% up to 40% of the "BMS SOC", and 3.28V up to 10% of the "BMS SOC", with an approximate average of 3.25V. This means that there is more usable capacity not communicated by the manufacturer. This silence is probably to mask the perceived 5% degradation in the first year that everyone is talking about.

In reality, does the battery really lose this 5%? Hard to prove without a full discharge followed by a full recharge. This is what I have done several times. After a year of use, my estimated SOH was down to 95%. By completely draining the battery and recharging it, I gained +0.80% each time, until I got this estimated SOH back up to 97.11%.

Even after a year and 23,000 km, I still discharge 50.80 kWh. Today, my car is 2 years old and 44,500 km, and I still discharge the same capacity. This is due to the generous usable capacity that was initially defined but not officially communicated.

Raw capacity revealed

When I go down to 0% displayed on the dashboard, this corresponds to a BMS SOC of 4.5%, which means that there is still energy left. At 0% BMS SOC, the cell voltage is 3.10 V, so the battery still has juice. By using consumers like defrosting and ventilation (about 750 W), I pushed the discharge to 2.88/2.99 V over a period of 2 hours. I thus recovered an additional 1.5 kWh.

Photo of a 0% dashboard discharge, after the dashboard counter has stopped counting

By fully recharging after this deep discharge, the electric meter recorded 59.2 kWh, and my terminal indicated 55.05 kWh injected into the battery at 32 A.

An LFP battery cannot be used at 100% of its gross capacity, as it is limited to about 87% to preserve its lifespan. The gross capacity of the MG4 LFP cells is 173Ah, or about 58.5kWh. This value can vary depending on the cells.

If we apply the 87% rule, this gives a theoretical usable capacity of about 50.89kWh. In my case, by pushing the discharges deeper (to 2.88v/2.99V), the CCU reset the zero point, allowing more capacity to be exploited. However, this setting reverts to 3.10V each time the CCU or BMS is updated at the dealer.

Therefore, no significant loss of autonomy due to estimated degradation is observed. Additionally, the recently announced 49kWh battery is simply the same battery as the 51kWh, with a more realistic advertised capacity based on the dischargeable energy between 100% and 0% of the dashboard (without considering the BMS SOC).

 

[EVentropy]

Wow, a lot of effort there. Bravo!
Always suspected 3.2V nominal was a generic figure for average voltage.
A bit of "fudging" or fuzzy logic going on to keep the battery out of trouble zone and user relatively
happy with performance.
Then there is what Terry said about nominal voltage being under "X" load, with a volt drop?
BYD LiFePO4 have different specs. A little higher.
I am hoping someone can corroborate my experience of 99% (100%) SOC being the triggered when the highest cell reaches 3.65V, and full power charging ceases/balancing starts.
Or is my bank an outlier?