SOC vs HV/LV delta and balancing….

[nanomad]

No need to guess, just use vdi and you can see what's data is available to service centers as well.

The MG4 BMS groups cells into 9 AFE. For each AFE you get min and max cell voltage plus some temperature stats.
If you get VDS you can even live log the data while driving and export it for later analysis if I remember right

 

[Romeo77]

Yep that all makes sense but why doesn’t MG4 “balancing facility” do that automatically when it hits 100%! Four hours at 100% should be plenty of time. Surely the balancing facility doesn’t shut off at a delta of 0.15V. And if so, what was the delta before it commenced balancing? I am leaning towards its is not enabled in software, or it is faulty. The reducing of the the high cell back into flat zone by use of the vehicle is irrelevant.
Unbelievable that there is no manufacturers documentation of this process.

If passive balancing is used, that takes a long time to correct any imbalance.
Not MG owner so can’t say what kind of balancing used, I own BYD, latest finding is that my car balancing works when battery discharges or charges with cycles of 17 hours from full charge, then after full charge and start driving continue to balancing cells for another 17 hours when car used.
But if i waiting for low SOC to start charge or charge once a week I almost always miss that balancing timing, and after reaching 17 hours limit my battery stays unbalanced till I charge it fully.
After i found out that i see a balancing improvement.
Have to say that is weird as if a driver like me has low mileage (fewer hours per week driving) and charge every day he is suffering from huge cell imbalance, same happens if using slowly battery over month to discharge it deep.
I found that by using a Launch X-431 CRP919 EV diagnostic tablet.
It supports almost all EV Brands (MG is supported) and can scan SAIC BMS to obtain full battery information too.

 

[EVentropy]

I’ve had one (VDI) in my shopping cart from OBd2shop for a month. Time to pull trigger.
And I’m going to have to upgrade my toughbook with XP on it I suppose. Another rabbit hole.
“Why did you buy that….. um, ……..work upgrade”
Rather have oem than launch, but launch sounds good…….

If passive balancing used that takes long time to correct imbalance.
Not MG owner so can’t say what kind of balancing used, I own BYD, latest founds is in my car balancing works when battery discharges or charges with cycles of 17 hours from full charge, then after full charge and start driving continue to balancing cells for another 17 hours when car used.
After i found out that i see balancing improvement.
Have to sai that is weird as if driver like me have low mileage (less hours per week driving) and charge every day he suffering from huge cell imbalance, same happens if using slowly battery over month to discharge it deep.
Found that by using Launch X-431 CRP919 EV diagnostic tablet.
It supports almost all EV Brands (MG supported) and can scan SAIC BMS to obtain full battery information to.

Can you post a screenshot of whole battery individual cell voltages please ?
Also while balancing, voltage delta vs time trace?
I would much appreciate it!

 

[Romeo77]

I’ve had one (VDI) in my shopping cart from OBd2shop for a month. Time to pull trigger.
And I’m going to have to upgrade my toughbook with XP on it I suppose. Another rabbit hole.
“Why did you buy that….. um, ……..work upgrade”
Rather have oem than launch, but launch sounds good…….


Can you post a screenshot of whole battery individual cell voltages please ?
Also while balancing, voltage delta vs time trace?
I would much appreciate it!

OEM diagnostic tablet for my car cost like whole car
screenshoot will not give you any usable info i think
my car is BYD Atto 3
but i can post diagnostic check result in pdf format.
"Also while balancing, voltage delta vs time trace" this will be different for different charging current and most of balancing in BYD done when battery discharges, OBC in my car not lowering charging current at 99% it keep charging untill any of cells get to 3,75 volt and then charging stops and battery capacity resets to 100%
If i want to balance during charging i need EVSE that can ask the OBC to draw 6-10 amps from wall socket or at least use type 2 cable 16 amper single phase (but car will draw 20 amps with this cable anyway)
In general charging options in BYD not perfect (current can`t be set in car infotainment. bad balancing strategy, not able to manualy preheat battery)
If BYD will not rectify that i will buy MG4, but in my country i can get only X Range model with NMC battery because LFP model with 51 kWh battery not popular (bad charging infransructure) and now importer not selling this model.
in attached file you will not see such high voltage but cells marked red is ones that get too high during charging.

 

[EVentropy]

The Atto3 is LiFePO4 right? If so, that is a high cell, high voltage cutoff for LiFe. 3.75V suggests its for NMC.

"OBC in my car not lowering charging current at 99% it keep charging untill any of cells get to 3,75 volt and then charging stops and battery capacity resets to 100%"

Similar to MG4. I just repeated this balancing check and video'd it. Not a lot to see just whats below.
Gets to 99%, HV cell trips simultaneously at 3.66V and charging goes to "trickle mode" on FICM and instantly reports 100%. Here I think (?) HV cell gets top trimmed by resistive balancing until trickle charge notation switches off on FICM at delta of ~150mV. Balancing top cell slows considerably until I provide a load, which could be just surface charge slowly blowing off.
Dollars to donuts there is no supportive charging to LV cell(s) to do a decent balance. Or trickle charging whilst burning off HV cell does very little. Not criticising just curious what is happening just not what my other LIFePO4 banks do.
Which is, balancing starts at 3.4V , with hi ---> low, while still maintaining decent charge rate so average bank voltage gets higher before "trickle charge" & resistive burn off. No point in starting it before 3.4V.

Essentially it means 100% SOC is governed by highest voltage cell and without active balancing results in well, not full capacity; the cheap way of doing it.

 

[wattmatters]

3.75V suggests its for NMC.

No. NMC is up to 4.3 V typically.

My 64 pack voltage at 100% SOC is 440 V, which for 104 cells is an average of 4.23 V.

Pack voltage last time I charged to 100% SOC:

 

[EVentropy]

Right! Good to know as I have no experience with NMC.
I would not be happy with a cutoff of 3.75V as factory specs for my other LiFePO4 is 3.65V max.
BYD builds their own batteries. Perhaps their specs are higher?

 

[wattmatters]

LiFePO4 can tolerate 3.8 V but it's not a recommended charge voltage - they are fully charged at 3.65 V. I charge my home LiFePO4 to an average of ~3.53 V.

 

[Romeo77]

The Atto3 is LiFePO4 right? If so, that is a high cell, high voltage cutoff for LiFe. 3.75V suggests its for NMC.

"OBC in my car not lowering charging current at 99% it keep charging until any of cells get to 3,75 volt and then charging stops and battery capacity resets to 100%"

Similar to MG4. I just repeated this balancing check and video'd it. Not a lot to see just whats below.
Gets to 99%, HV cell trips simultaneously at 3.66V and charging goes to "trickle mode" on FICM and instantly reports 100%. Here I think (?) HV cell gets top trimmed by resistive balancing until trickle charge notation switches off on FICM at delta of ~150mV. Balancing top cell slows considerably until I provide a load, which could be just surface charge slowly blowing off.
Dollars to donuts there is no supportive charging to LV cell(s) to do a decent balance. Or trickle charging whilst burning off HV cell does very little. Not criticising just curious what is happening just not what my other LIFePO4 banks do.
Which is, balancing starts at 3.4V , with hi ---> low, while still maintaining decent charge rate so average bank voltage gets higher before "trickle charge" & resistive burn off. No point in starting it before 3.4V.

Essentially it means 100% SOC is governed by highest voltage cell and without active balancing results in well, not full capacity; the cheap way of doing it.

There is a difference between balancing strategies on these two brands !!!
On my Atto 3, OBC uses constant power, all it cat get from power source, until “100%” SOC - i.e. until any of the 126 cells reach HVD set voltage of 3.75 V.
Then the BMS starts to extra discharge HV cells during usage of the car, and if I start to charge the battery in the period 17 hours after reaching 100% the balancers are working also when I charge the car, all the time until the 17 hours limit is reached….
Then cycle repeats after reaching 100% SOC again.
On MG, during balancing stage when you see low power draw, discharging of HV cells occurs and at the same time LV cells are getting a low power charge.
I can simulate same balancing stage on my car but to do that I need during charging to add a load to lower current reaching battery, so that net current is not higher than balancing current.
I did that once, and saw a better balanced battery, but that is too complicated, since my car does not have a PTC heater and the heat pump power varies its power….

[ Edit moderator: I heavily edited the above for readability; the poster's first language is obviously not English. I hope that I have not mangled the meaning. No criticism of the post is implied. ]

~60%, bit scarey, but bearing in mind that was me flooring it and did return to 3.3Vish instantly load was off and regen was no longer happening.
Curious to know what is trigger for limp mode?
Min cell voltage or what? Also going to 0% I doubt will work for me as I have a steep 100m driveway to garage.
Cheers.

Think that there no difference between brands of car, for BYD LFP is 90 mV delta at 50% SOC
Then it limits power …..

I charge our MG4 SE SR with a granny charger. The mains supply to the charger is switched on with a Shelly relay and this also monitors the mains power. And I can plot this with Home Assistant automation.
Not sure if this helps but I've just finished a balancing charge and this is what the mains power looks like:

View attachment 33501

Interesting
That looks exactly like in case of BYD Atto 3 with passive balancing !!!
But my Atto 3 not doing that automatically
To achieve this I need to start delay change in car and connect it to charger, when car wait charge time the BMS balancing cells, what i see exactly same, BMS letting to first cell reach defined value then stop supply current to battery and discharging cells that have too high voltage.
Then topping up and repeat discharging.

 

[EVentropy]

Think that there no difference between brands of car, for BYD LFP is 90 mV delta at 50% SOC
Then it limits power …..

Did you do that in real time with carscanner accelerating hard until you got a delta of 90mV @50% soc and then flooring it, or is that from manufacturer?
Don't laugh, I have done similar!

 

[Romeo77]

Did you do that in real time with carscanner accelerating hard until you got a delta of 90mV @50% soc and then flooring it, or is that from manufacturer?
Don't laugh, I have done similar!

This from repair manual for my car about 90mV at 50% , wish to everyone not to reach such situation !!!
Meanwhile my battery balancing cells better every cycle, who that idiot that did such dumb balancing procedure in BYD
Now after i found how it works i see 123 mV at moment of finish changing.
Cost me buying Launch X-431 diagnostic….
Month before delta fluctuated 360 mV to 250 mV

@EVentropy
Just out of curiosity
How much time your MG4 spend powered on between charging to 100%
Why i ask?
Because when i got access to charger at work i started charging every day without enough time of car powered on (driving only 15 minutes to work and back to home)
On Atto 3 balancing work mostly when we drive car or sitting in it when it powered on.
Then also during charging (not at the end of charging) But balancing time limited for some reason - only works 17 hours from time of full charge.
Also stopping balancing if we charge by using DC rapid station.

 

[EVentropy]

Now after i found how it works i see 123 mV at moment of finish changing.
Cost me buying Launch X-431 diagnostic….
Month before delta fluctuated 360 mV to 250 mV

That is 123mV moment of finish charging or charging and balancing?
That is, exactly when it hits100% or a time period after when balancing is complete?

Your balancing method seems to be working well.
I googled your model an it appears LiFe blade batteries max cutoff is 3.8V as opposed 3.65V for the MG. And maybe my car does not reach first parameter for stopping charging and using BMS max cell voltage as trigger to trickle charge simultaneously with balancing.

I regard the bench mark delta to be when vehicle is at rest with no load or only very small load so results can be standardised. I assume Atto 3 balancing starts at 100% like MG4?

What I understand you to say is when you hit 100% your delta (previous to your new balance method) was 250 to 360mV now is only 123mV before balancing! If so, that is a excellent result and what I was trying to do. Even the cells out!!
You also say the best way to balance is driving the car and then delta drop REAL quick to delta < 20mV (no load) Totally agree!!
I have a 1000' hill close by so I can blow off the delta real quick but have to turn regen off on way down and use the brakes which I do not like. Then I could do repetitive charges to bring bottom cells up.

Just out of curiosity
How much time your MG4 spend powered on between charging to 100%

Max 10mins after reaching 99% then jumps to 100% immediately and approx 5-10mins of relay clicking in back with power from AC going from 0 to 1.8 kW

 

[Romeo77]

That is 123mV moment of finish charging or charging and balancing?
That is, exactly when it hits100% or a time period after when balancing is complete?

Your balancing method seems to be working well.
I googled your model an it appears LiFe blade batteries max cutoff is 3.8V as opposed 3.65V for the MG. And maybe my car does not reach first parameter for stopping charging and using BMS max cell voltage as trigger to trickle charge simultaneously with balancing.

I regard the bench mark delta to be when vehicle is at rest with no load or only very small load so results can be standardised. I assume Atto 3 balancing starts at 100% like MG4?

What I understand you to say is when you hit 100% your delta (previous to your new balance method) was 250 to 360mV now is only 123mV before balancing! If so, that is a excellent result and what I was trying to do. Even the cells out!!
You also say the best way to balance is driving the car and then delta drop REAL quick to delta < 20mV (no load) Totally agree!!
I have a 1000' hill close by so I can blow off the delta real quick but have to turn regen off on way down and use the brakes which I do not like. Then I could do repetitive charges to bring bottom cells up.

Max 10mins after reaching 99% then jumps to 100% immediately and approx 5-10mins of relay clicking in back with power from AC going from 0 to 1.8 kW

That delta i catch at the moment of charging stops (and that delta also stored in BMS for each full charge cycle), then it falls quickly to 87 now, and after 10 minutes of car started it’s already 33 mV, nope BYD balancing cells also when discharge of battery happens in time frame of 17 hours from previous 100% charge achieved, and we not have clear indication of balancing other then charging time prediction longer than it usually should be.
Delta drops when we start discharging battery only because cells that gets to 3.75 volts (in my case) then drops quickly to rest voltage of 3.3 approximately.
That not caused by balancing.
Might be CATL also balancing cells when battery connected i.e. car powered on in some way ?
Try to configure delayed charging in car infotainment before charging starts to lets say 2 hours after you arrive home, and at that time watch delta, if you will see delta fluctuations when car awaiting charging time then it definitely balancing while car powered on, and not only when charging finished.
If after several such cycles you will see improvement (lowest cell voltage get closer value to cell with high voltage) than we found right way to get battery pack balanced also in MG.