Project OBD2 - Mobile App v1.2.1 and later

well, I can understand why it says 9x% rather than 100% - as we know we're not allowed to use the full 44.5wKh battery capacity.

This isn't a figure for people to see, so it makes sense that it maybe the real internal SoC of the whole 44.5kWh capacity.

The BMS etc then will determine I'd imagine how much to allow you to charge and discharge to based on this percentage of the actual whole 44.5kWh capacity. Further, based on the amount of the whole 44.5kWh capacity we're allowed to use, they show us the percentage on the dash of the amount we're allowed to use.

E.g. They say we're only allowed to charge and discharge to 95% & 5% of the battery capacity.
So when charging it stops at 95% of 44.5kWh, but they show that as 100% to the user on the dash. When discharging it stop at 5% of 44.5kWh, but they show that as 0% to the user on the dash.

If I were designing the system, that it probably how I'd do it, it is logical.

Re SOH - I have no idea how this is determined, I do not have enough knowledge of how degradation of batteries are measured. Some further night-time reading is needed :)
There’s a top paid job looming for you in the EV industry
 
So can we rely on what the app is telling us?
I think if soneone with OVMS and some minor degradation, can check with this app then we will might have better idea of the accuracy of the app.

If your at the 1 year mark, at the service, did they give you the service check-sheet (prev posted somewhere in the forum) where SoH documented, if so does it agree with the app?
 
Last edited:
I think if soneone with OVMS and some minor degradation, can check with this app then we will might have better idea of the accuracy of the app.

If your at the 1 year mark, at the service, did they give you the service check-sheet (prev posted somewhere in the forum) where SoH documented, if so does it agree with the app?
The car has been serviced, but I only had a standard service record and nothing recorded or noted in relation to the battery degradation.
I would be presently surprised if this were the case, that said I’ve done very little rapid charging maybe 7 or 8 times...
 
I've been thinking and reading and, have come up with the following, enjoy!!!

Battery State of Charge & State of Health hypothesis.

Links:
State of charge
https://www.analog.com/media/en/tec...ge-and-state-health-estimation-techniques.pdf
https://res.mdpi.com/d_attachment/wevj/wevj-11-00066/article_deploy/wevj-11-00066.pdf

The above reference are quite informative regarding ways of calculating the State of Charge & Health of a battery.

There are several ways of calculating it. I am going to presume and base the following hypothesis on MG using the “simpler” method which is based upon Voltage and the known discharge curve of the cells being used.

To confirm, this is just my hypothesis based on current information.

State of Charge:
  1. There is a known discharge curve for the battery cells being used in the MG ZS EV. From this, with a given voltage one can calculate the State of Charge of the battery.
  2. The specified maximum charge voltage per battery cell X 108 cells is what will be the Total Maximum Charge Voltage (e.g. 459v) and will be considered the point at which the battery is fully charged (and is storing 44.5kWh of power), referred to as 100% ‘REAL State of Charge’.
  3. The specified minimum discharge cut-off voltage per battery cell X 108 cells is what will be the Total Minimum Discharge Cut-off Voltage (e.g. 324v) and will be considered the point at which the battery is fully discharged (and is storing 0.0kWh of power), referred to as 0% ‘REAL State of Charge’.
  4. The voltage Discharge curve for the battery cell is used to calculate the REAL State of Charge % between the aforementioned Max & Min, it is very much NOT a linear line.
  5. To avoid over stressing the battery cells, the BMS is configured to not allow a 100% REAL charge and is not allowed to discharge to 0% REAL charge.
  6. If BMS is configured to keep the REAL State of Charge between 2% & 97% (giving a usable 42.3kWh out of 44.5kWh).
    From a users perspective, it will treat/display this as what is the 0% & 100% Usable State of Charge.
    The Formula would be: Users Usable State of Charge % = ( [REAL State of Charge %] - 2 ) / 0.95
    The BMS configuration figures are not actually known, but we know have changed in BMS versions.

State of Health I am going to propose a guess is as follows:
  1. The battery has a 100% REAL State of Health when it can store it’s full specified maximum amount of power - 44.5kWh.
  2. Fundamentally, the REAL State of Health % = the REAL State of Charge %.
  3. The BMS will not allow the battery to charge to more than a real 97%, as such the battery will never receive a full charge, so one will never see a real 100% State of Health.
  4. It cannot be said that a new battery does not have less than a 100% State of Health, therefore a REAL State of Charge of 97% will be treated as 100% SoH. It is kind of true to say that the battery is fully working if it can be charged to what it ever wants to charge it to - 97%.
  5. IF [REAL State of Charge] > 97% THEN State of Health = 100% ELSE State of Health = [REAL State of Charge]
  6. This would mean that no one will see a State of Health less than 100% UNTIL it gets down to 96%, which could explain why no one yet seems to have seen a non 100% SoH - one would expect some cars to be a bit less than 100% SoH by now.
  7. The BMS configuration figures are not actually known, but have changed in BMS versions.

Does the Users Usable State of Charge dynamically alter based on State of Health:
  1. If it were to not dynamically change, once a reasonable amount of battery degradation occurs, the user would never see the car charge to “100%”.
    This would cause issues such as:
    When charging, the user would never know or be able to predict when the car would be at it’s fullest possible state of charge - as the target % would be a moving target.
    Should the user, for example, want to charge the battery to 80%, but the State of Health is 75%, it will never get to 80% - the actual target for 80% of the current possible charge would be 60% (80% of 75%).
  2. The Users Usable State of Charge % could be dynamically calculated:
    By treating 100%, as the State of Health % instead of the BMS Maximum configured (e.g. 97%).
  3. Until we see cars with degraded batteries (SoH), we won’t know if it dynamically alters the Users Usable State of Charge.
 
That's a nice high voltage you've got yourself there sir!

- I've added this full charge voltage and ECU SoC data to the spreadsheet.
Yes 🙂 Now we have the App (which we all hope is reasonable accurate) and your excellent spreadsheet it's good to have a reference point that as all our car's milage and age increases we can visually see any changes that may occur and also how any further BMS updates alter the HV battery peramiters 👍
 
Screenshot_20210306_112705.jpg

Less than 100% SOH
8-9K mileage.

Edit. User is @Freddie
R17 BMS... ?? Wasn't that the original
Possible ECU SoH has been manipulated on later updates... Will be good to compare after an update
Screenshot_20210306-103357.png
 
Last edited:
Less than 100% SOH
8-9K mileage.

Edit. User is @Freddie
R17 BMS... ?? Wasn't that the original
Possible ECU SoH has been manipulated on later updates... Will be good to compare after an update
I just saw this on other post.

The first non 100% SoH (although Mike has it too via OMVS he says).

Yeah R17 is a very old BMS, not the very original I believe but the 1/2020aka12/2019 version.
I was thinking exactly the same thing, have they changed the way it's calculated....
 
To confirm, you charged to 100% on a rapid charger, then took the reading - 97% on app? Did you get the Voltage and min & max cell voltages (would show if out of balance much/at all).

You have the 455V BMS, like mystreet & Tango130 - they're showing 96% ECU SoC.
Most of my charging so far has been on rapids. Only had one or two charges on 7kW, and not all of those were long enough to get to balancing. 0.014V variance is well within tolerance.
 

Attachments

  • Screenshot_20210302-162840_ZS EV.jpg
    Screenshot_20210302-162840_ZS EV.jpg
    107.1 KB · Views: 158
Most of my charging so far has been on rapids. Only had one or two charges on 7kW, and not all of those were long enough to get to balancing. 0.014V variance is well within tolerance.
WOW 457.2 VOLTS

I don't know what is actually considered good tolerances with regard to the min & max cell voltage. Agreed 0.014 seems pretty good - Mine is usually about that and I always AC charge.
 
Full battery is 97% in the App.
Guessing top and bottom buffer 1.5%.

If the January update also get 97% then will confirm available power range is the same and likely buffer percentages were indeed changed.

Equalising/balancing current is around minus 0.2A.
I don’t know if this has been answered, there are too many pages to look through I gave up the will.

OVMS also reported 96/97% when fully charged on the old software. I believe that it now has a 6% top buffer and 3% at the bottom. MG just shifted from 3% at the top & 6% at the bottom. Gives more room for regen at 100% where it is more useful and helps protect the battery more plus you get faster charging at a rapid as well as the cell voltages are lower. So a 9% buffer in total out of the 44.5kWh, so a useable capacity of roughly 40.5kWh.
 
I don’t know if this has been answered, there are too many pages to look through I gave up the will.

OVMS also reported 96/97% when fully charged on the old software. I believe that it now has a 6% top buffer and 3% at the bottom. MG just shifted from 3% at the top & 6% at the bottom. Gives more room for regen at 100% where it is more useful and helps protect the battery more plus you get faster charging at a rapid as well as the cell voltages are lower. So a 9% buffer in total out of the 44.5kWh, so a useable capacity of roughly 40.5kWh.

It's good to have the confirmation that OVMS was reporting the same.

Personally I'm abit skeptical as to why the buffers were changed its possible that the larger top buffer is there to hide some battery degradation...I have heard Tesla did a similar thing at some point.
It's also possible that both companies recognise it's better for battery longevity.
The other side effects that you mentioned are very welcomed though.
Did you have a SOH reading before and after the mid-Jan BMS update
 
It's good to have the confirmation that OVMS was reporting the same.

Personally I'm abit skeptical as to why the buffers were changed its possible that the larger top buffer is there to hide some battery degradation...I have heard Tesla did a similar thing at some point.
It's also possible that both companies recognise it's better for battery longevity.
The other side effects that you mentioned are very welcomed though.
I don’t think there is anything sinister going on. Protecting the battery, giving us more regen where it is useful and faster charging speeds are a definite bonus. We haven’t lost range or capacity and my SoH did not change from before the update.
 
Here's mine
 

Attachments

  • Screenshot_20210306-211946_ZS EV.jpg
    Screenshot_20210306-211946_ZS EV.jpg
    165.7 KB · Views: 135
  • Screenshot_20210306-211841_ZS EV.jpg
    Screenshot_20210306-211841_ZS EV.jpg
    113.6 KB · Views: 132
  • Screenshot_20210306-211828_ZS EV.jpg
    Screenshot_20210306-211828_ZS EV.jpg
    108.8 KB · Views: 131
  • Screenshot_20210306-211817_ZS EV.jpg
    Screenshot_20210306-211817_ZS EV.jpg
    144.8 KB · Views: 130
Support us by becoming a Premium Member

Latest MG EVs video

MG3 Hybrid+ & Cyberster Configurator News + hot topics from the MG EVs forums
Subscribe to our YouTube channel
Back
Top Bottom