Is this SoH OK?

[decrep]

I think I get the idea that you can bleed off excess energy of individual cells when charging has concluded.

What I’m struggling with is the concept that MG has found a (low cost) way to discharge high voltage cells while simultaneously topping up the rest in order to achieve equilibrium??

As coulomb described, there are bleed resistors across the high voltage cells to discharge them, while there is still a small charging current going into all cells. I think this is why the charging current drops right down while balancing. The bleed resistors are carrying more than the charge current, other wise it wouldn't work.

 

[T1 Terry]

I think I get the idea that you can bleed off excess energy of individual cells when charging has concluded.

What I’m struggling with is the concept that MG has found a (low cost) way to discharge high voltage cells while simultaneously topping up the rest in order to achieve equilibrium??

The problem is, these bleed down resistor are such a low value that levels like 2milliamp discharging is about the max.
These resistors turn on when a cell goes high voltage, but 4/5ths of FA reduction in charge current to the high cell resulting in the high cell going even higher till something trips the cell over voltage stop charge.
From that point, the low cell remains low and the capacity from the high cell/s is wasted as heat and this will require cooling, wasting more energy from all the cells if the charger doesn't supply it.

Each repeat balancing operation will get the low cell higher, until eventually (it is hoped) the full pack end of voltage occurs before a cell goes over voltage and stops the charge.
From that point on, every charge to 100% will get the battery closer to fully balanced

We live in hope that a BMS manufacturer will build an active balancing unit that manufacturers will adopt and save all this nonsense and rectify the declining battery health problem ......
At the moment, about the best solution available is 7 x 16 cell units to allow an overlap between the units to move a high cell voltage progressively through the units to reach a low cell that is out of range of the single 16 cell active balancer unit.
Another company, Deligreen Li-ion Battery Equalizer Balancer have individual cell balancers that connect together in a chain to move the capacity from the high cell to the low cell

T1 Terry

 

[Data]

We shouldn't forget that 'Passive Equalisation' works very well. That's why pretty much everyone uses it. Active balancing/equalisation is no doubt very slightly better but the advantages of that are really not worth the very large increase in cost.

 

[wattmatters]

We shouldn't forget that 'Passive Equalisation' works very well.

Especially if the cells and packs are all really well matched to begin with. With lower grade cells that are not well matched, then active balancing is going to be required.

 

[Data]

Thankfully MG batteries are excellent quality with very few issues relative to the number of cars out there. Like Tesla they tend to run their cells 'under volted' so they are never under a lot of stress.

 

[wattmatters]

As coulomb described, there are bleed resistors across the high voltage cells to discharge them, while there is still a small charging current going into all cells. I think this is why the charging current drops right down while balancing. The bleed resistors are carrying more than the charge current, other wise it wouldn't work.

The resistors can also bleed energy from a cell when the pack is not being charged, so with balance charging switching on/off or ramping up and down in power delivered this can effectively result in a cell with high voltage reach a quasi steady state voltage or even drop down while over time the rest of the cells' voltages catch up. It's just a matter of how that power cycling is managed. And of course which cell is "highest" often changes through this process. Sometimes more than one cell is chosen to be discharged, all depends on the BMS programming and circuitry used.

The actual power delivery during an equalisation charge session is not constant but rather jumps up and down. e.g this is the power demand from the charger during an equalisation phase at the end of a full 100% charge:

The equalisation phase on this day took over and hour and half. In the above I am only capturing power delivery changes at one minute average power level, so shorter duration variation is masked.

This is the zoomed out chart showing the slow charge leading up to reaching what the BMS thinks is 100% SOC, but there was still room for more charge before then moving to equalisation phase:

 

[Everest]

What I’m struggling with is the concept that MG has found a (low cost) way to discharge high voltage cells while simultaneously topping up the rest in order to achieve equilibrium??

I can understand the confusion. The high voltage cells are not being discharged. They are just being charged at a slower rate than the others to minimise the voltage delta (as T1 explained above) - see my diagram below for a (hopefully) clearer explanation.

As coulomb described, there are bleed resistors across the high voltage cells to discharge them, while there is still a small charging current going into all cells.

almost... but not quite

The bleed resistors are carrying more than the charge current, other wise it wouldn't work.

No... that is impossible in a series circuit that is being charged.

As @Coulomb explained, the BMS will switch a 'bleed' resistor in parallel with one or more 'higher-voltage' cells during charging.

Thanks to Kirchoff's current law, the total current entering a node must equal the total current leaving it. So in the diagram below Icharge = Icell + Ibleed.

When a cell reaches a higher voltage than the others, the BMS activates a solid state switch that connects a shunt/bleed resistor in parallel with that specific cell.

Because a portion of the current is diverted through the bleed resistor the current entering the 'highest-voltage' cell is reduced, allowing other lower-charged cells to catch up hopefully before the high-voltage cell reaches its peak as some current ( the I-cell element) will still be charging it.

 

[wattmatters]

No... that is impossible in a series circuit that is being charged.

I think they meant more than what is going through that cell, i.e. siphoning off power as per your diagram.

 

[Coulomb]

The bleed resistors are carrying more than the charge current, other wise it wouldn't work.

No, it can still work, as long as the "unbalance" isn't too great. All cells will charge, but the ones that have the highest voltage will charge slower than the others.

If one of the cells gets full despite the bleed current, the charge current can be reduced to reign it in. In the worst case, that charge current will be less than the bleed current, so that the highest voltage cell can be drained.

Edit: Oops, I see that this has already been covered.

 

[T1 Terry]

We shouldn't forget that 'Passive Equalisation' works very well. That's why pretty much everyone uses it. Active balancing/equalisation is no doubt very slightly better but the advantages of that are really not worth the very large increase in cost.

Just because it's cheap and easy to build, doesn't mean it works well, because it doesn't ... never has and never will, but it's what we have, so we have to work with it ......

T1 Terry

 

[Data]

Just because it's cheap and easy to build, doesn't mean it works well, because it doesn't ... never has and never will, but it's what we have, so we have to work with it ......

T1 Terry

My point being that very few people actually get issues with batteries. It is clear "it works ok" might be a better way of putting it. We know active balancing is better, but not really needed.

 

[T1 Terry]

Thankfully MG batteries are excellent quality with very few issues relative to the number of cars out there. Like Tesla they tend to run their cells 'under volted' so they are never under a lot of stress.

That was Toyota's logic with the NiMh hybrid battery packs, keep a big safety margin at the top and bottom of the voltage range and use 3Ah out of a 6Ah battery .....
Yes, it does delay the eventual battery failure, but unless the user goes through the deep discharge and slow recharge to 100% on a frequent basis, the battery SoH continues to fall, resulting in reduced range from a full charge .....

T1 Terry

 

[T1 Terry]

I can understand the confusion. The high voltage cells are not being discharged. They are just being charged at a slower rate than the others to minimise the voltage delta (as T1 explained above) - see my diagram below for a (hopefully) clearer explanation.


almost... but not quite


No... that is impossible in a series circuit that is being charged.

As @Coulomb explained, the BMS will switch a 'bleed' resistor in parallel with one or more 'higher-voltage' cells during charging.

Thanks to Kirchoff's current law, the total current entering a node must equal the total current leaving it. So in the diagram below Icharge = Icell + Ibleed.

When a cell reaches a higher voltage than the others, the BMS activates a solid state switch that connects a shunt/bleed resistor in parallel with that specific cell.

Because a portion of the current is diverted through the bleed resistor the current entering the 'highest-voltage' cell is reduced, allowing other lower-charged cells to catch up hopefully before the high-voltage cell reaches its peak as some current ( the I-cell element) will still be charging it.

View attachment 44697

Looks good on paper, but the resistor is actually controlling the current in a short across the cell to burn off capacity like a mini electric radiator.
The cell top boards used in the Blade electron actually had two ceramic resistors mounted clear of the board to help dissipate the heat while it tried to bring the high cells down.
For such old technology, the system actually allowed enough safety margin for the cells to still be driving the vehicle after 10 yrs or more.

T1 Terry

 

[Data]

That was Toyota's logic with the NiMh hybrid battery packs, keep a big safety margin at the top and bottom of the voltage range and use 3Ah out of a 6Ah battery .....
Yes, it does delay the eventual battery failure, but unless the user goes through the deep discharge and slow recharge to 100% on a frequent basis, the battery SoH continues to fall, resulting in reduced range from a full charge .....

T1 Terry

Yes indeed, but it works. Look at how many very high mileage Tesla's & indeed MG's there are out there. Batteries still in very good condition. As per our owners manuals, we all need to carryout full & proper equalisation charging to maintain our batteries & give them a long life whether LFP or NMC. Many don't do this & wonder why they are losing range.

Of course many vehicles try to do the equalisation charging without owner intervention. Even our MG's try to do this (but it's pretty much last resort). However many have found this often doesn't work as they are using 'Smart Charging' which doesn't allow equalisation to happen. It simply shuts off at 100%. My friends Polestar 2 tries to do this but fails. Therefore he didn't know he was supposed to do equalisation charging and had been complaining of his range dropping steadily. I checked his battery & it was very unbalanced. After a couple of balance charges he now has restored range & SOH is still very good at 98%. It was 97% prior to equalising. Mileage is 30,000 or so. Car is two years old. I think the biggest issue regarding battery health is owners not quite being up to speed with how to look after them. Few people are experts on this, but we shouldn't have to be. Manufacturers could be clearer in their manuals concerning this aspect of owning an EV.

 

[T1 Terry]

Yes indeed, but it works. Look at how many very high mileage Tesla's & indeed MG's there are out there. Batteries still in very good condition. As per our owners manuals, we all need to carryout full & proper equalisation charging to maintain our batteries & give them a long life whether LFP or NMC. Many don't do this & wonder why they are losing range.

Of course many vehicles try to do the equalisation charging without owner intervention. Even our MG's try to do this (but it's pretty much last resort). However many have found this often doesn't work as they are using 'Smart Charging' which doesn't allow equalisation to happen. It simply shuts off at 100%. My friends Polestar 2 tries to do this but fails. Therefore he didn't know he was supposed to do equalisation charging and had been complaining of his range dropping steadily. I checked his battery & it was very unbalanced. After a couple of balance charges he now has restored range & SOH is still very good at 98%. It was 97% prior to equalising. Mileage is 30,000 or so. Car is two years old. I think the biggest issue regarding battery health is owners not quite being up to speed with how to look after them. Few people are experts on this, but we shouldn't have to be. Manufacturers could be clearer in their manuals concerning this aspect of owning an EV.

The charging doesn't stop at 100%, it stops because a cell went over voltage .... credit where credit is due, this is real plus for extended cell life.
If even the passive balancing was to start at say 3.45v and cut off at 3.4v, some balancing would occur every full charge ....
Unfortunately, many people think they are doing the right thing by their LFP cells by not bringing them up to 100% charged frequently .... it might be a necessity for NMC chemistry, but hopefully that will soon be phased out and replaced with the better LFP or Na+ chemistry .....

T1 Terry

 

[Data]

The charging doesn't stop at 100%, it stops because a cell went over voltage .... credit where credit is due, this is real plus for extended cell life.
If even the passive balancing was to start at say 3.45v and cut off at 3.4v, some balancing would occur every full charge ....
Unfortunately, many people think they are doing the right thing by their LFP cells by not bringing them up to 100% charged frequently .... it might be a necessity for NMC chemistry, but hopefully that will soon be phased out and replaced with the better LFP or Na+ chemistry .....

T1 Terry

Yes, understood. However for most people smart charging doesn't allow equalisation to happen as the cars BMS is not in charge. The electricity supplier is. As soon as their server sees your car is at 100% & the car is full, it shuts down the charge before equalisation occurs. Switching off smart charging puts the BMS in charge allowing it to control the charge all the way. Not just that, smart charging often interrupts the charge cycle. Octopus Energy (a great UK supplier) often charges your car in several isolated bursts. Rarely one long charge as required by most EV's needing a full equalisation. My own ZS EV requires a single long AC uninterrupted charge with no smart charge enabled. It will then equalise fully. Some on here have claimed they can equalise with smart charging switched on. I've not yet met anyone who can do that so for me that remains unverified. I guess it may be possible depending on whether the car or the wall box is listed as the first device in the configuration app. We have 5 EV's in the family to include VW ID4, E-Niros, Kona, Leafs & my MG. As we have discovered all require smart charging to be switched off to allow equalisation.

 

[T1 Terry]

Yes, understood. However for most people smart charging doesn't allow equalisation to happen as the cars BMS is not in charge. The electricity supplier is. As soon as their server sees your car is at 100% & the car is full, it shuts down the charge before equalisation occurs. Switching off smart charging puts the BMS in charge allowing it to control the charge all the way. Not just that, smart charging often interrupts the charge cycle. Octopus Energy (a great UK supplier) often charges your car in several isolated bursts. Rarely one long charge as required by most EV's needing a full equalisation. My own ZS EV requires a single long AC uninterrupted charge with no smart charge enabled. It will then equalise fully. Some on here have claimed they can equalise with smart charging switched on. I've not yet met anyone who can do that so for me that remains unverified. I guess it may be possible depending on whether the car or the wall box is listed as the first device in the configuration app. We have 5 EV's in the family to include VW ID4, E-Niros, Kona, Leafs & my MG. As we have discovered all require smart charging to be switched off to allow equalisation.

What causes the smart charging to believe the car has reached 100% SOC and to stop the charging?

The BMS should not require the charging to be continuous to do a top balance, but even using passive resistor type balancing, slowing the charge rate to equal to or less than the balancing resistor discharge capacity until all cells reach the 3.5 volt mark would still bring the pack to a proper top balance ......

The reason they don't do this is because of the heat the resistors generate would burn out the balance boards, plus the waste of electricity keeping the whole system powered for hrs would not be acceptable to most EV owners.

Active balancing would solve all of these problems, unfortunately, manufacturers are not willing to add the complexity or cost, when cheap and simply with big safe guards at each end of cell voltage will get them out of warranty problems ....

After all, is it in their best interest to have batteries that last a long time and never seem to degrade? Not the way to sell the customer a replacement vehicle is it?

T1 Terry

 

[Data]

What causes the smart charging to believe the car has reached 100% SOC and to stop the charging?

The BMS should not require the charging to be continuous to do a top balance, but even using passive resistor type balancing, slowing the charge rate to equal to or less than the balancing resistor discharge capacity until all cells reach the 3.5 volt mark would still bring the pack to a proper top balance ......

The reason they don't do this is because of the heat the resistors generate would burn out the balance boards, plus the waste of electricity keeping the whole system powered for hrs would not be acceptable to most EV owners.

Active balancing would solve all of these problems, unfortunately, manufacturers are not willing to add the complexity or cost, when cheap and simply with big safe guards at each end of cell voltage will get them out of warranty problems ....

After all, is it in their best interest to have batteries that last a long time and never seem to degrade? Not the way to sell the customer a replacement vehicle is it?

T1 Terry

This has been covered previously in other threads but always worth repeating.

The BMS in our MG's has its own program to allow it to conduct a 'full & proper equalisation & BMS Calibration'. The car ha to be in charge of equalisation & calibration. Smart charging cannot do this for your car. To allow it to run fully certain criteria has to be met.

These are (as per MG owners manual, MG dealer & MG customer support instructions):

• battery state of charge must be run down to when the 'low battery' warning sounds;
• smart charging must be turned off (it doesn't mention this in the manual);
• charging must be slow (granny or 7kw wall box) & continuous, no interruptions.

If the charging is interrupted the BMS program ceases to carryout a full & proper equalisation even when charging is restarted. Yes, I and others on here have tested this. It doesn't. You have to start again by driving the car to get the low warning again. Simples really.

Some folks forget to turn off smart charging prior to the equalisation charge. Therefore in nearly all cases the charge will be interrupted & in any case their electricity supplier will stop the charge at 100% before a full & proper equalisation takes place.

Worth noting a partial equalisation car be carried out at 70 or 80% or other soc's. This isn't a full & proper equalisation but clearly helps to keep the cells balanced to some extent.

 

[T1 Terry]

I think you misunderstood what I was getting at. There is no reason for the BMS to require a constant charge, it is just the way MG have set up their BMS, but more likely, the BMS system just about all EV makers use, why reinvent the wheel, even if isn't round it still sort of works.
The reason for the deep discharge is also reset the low cell voltage sensing, to ensure the battery only discharges as far as the cell with the least capacity.
If it's really smart, it will record which cell reaches minimum voltage and first and compare that to which cell reaches max voltage first, that way it would know which cell/s had lost capacity compared to the others ....
I have a feeling that is a level of sophistication isn't something the cheaper end of the market would be interested in measuring and most definitely not recording ant where in the onboard computer, for fear some smarty would figure out how to access it and use it against them on a warranty claim .....

T1 Terry

 

[Data]

I think you misunderstood what I was getting at. There is no reason for the BMS to require a constant charge, it is just the way MG have set up their BMS, but more likely, the BMS system just about all EV makers use, why reinvent the wheel, even if isn't round it still sort of works.
The reason for the deep discharge is also reset the low cell voltage sensing, to ensure the battery only discharges as far as the cell with the least capacity.
If it's really smart, it will record which cell reaches minimum voltage and first and compare that to which cell reaches max voltage first, that way it would know which cell/s had lost capacity compared to the others ....
I have a feeling that is a level of sophistication isn't something the cheaper end of the market would be interested in measuring and most definitely not recording ant where in the onboard computer, for fear some smarty would figure out how to access it and use it against them on a warranty claim .....

T1 Terry

Terry, yes of course to what you are saying. This is just how mostly all EV's are set up. Without the deeper discharge & a slow full charge to 100%, without smart switched on, a full & proper equalisation/balance isn't possible. It says that in our MG manuals (for the ZS page 9) & in every manufacturers car owners manual for just about every other EV that I've read.

I'm a retired auto tech. Although I have spent 18mths studying EV batteries I am not a battery eggspurt!! I do however know several people who are involved in EV battery research & development. It seems the long continuous charge provides superior accuracy with, amongst other things, a very stable temperature etc etc. And so that's why our cars BMS setup is as it is. Especially in winter it can make a bigger difference to how well the battery equalises. Apparently many BMS will record the cells that reach min & max voltages first etc. But it's really only info that's available to manufacturers or dealers. Our little OBD readers won't give us that. Whether MG BMS will record that is anyone's guess but the batteries are high quality so I wouldn't be surprised if they did.

I came to this excellent site when I bought my EV to get info on how to do a proper equalisation. I was hit with a raft of well meaning but totally incorrect information. None of which allowed me to do an equalisation. The MG owners manual isn't written for people to easily understand this process. I ended up having to contact MG for clarification, my dealer, and reading copious amounts of technical papers to understand the process. I experimented on all our family EV cars. I didn't need to do any of that but my engineering brain always has to know with certainty. I was shouted down on this site by some people who actually didn't seem to always know what they were talking about. But as an owner I feel I don't now need any further information. Terry, you, Everest, wattmatters & one or two others in this thread are very knowledgeable. In fact some of the most knowledgeable on the site concerning EV batteries. Thanks for your input.