MG4 Equalisation Charge

[Sloucher]

I have the 72 plate mg5 long range. Would putting the car on a Tesco 7kwh type 2 for couple of hours would this balance the battery if taken to 100% from about 80%

I'm not familiar with the battery chemistry of the MG5 I'm afraid. However, if it is, as I suspect, NMC
then I'm not sure a couple of hours would be long enough. It would depend on the condition of your battery in the first place.
First of all you need to get it to 100% and that last 10% does take quite a long time, then you have to leave it at 100% for a further period. I'm not sure other Tesco users would be too happy with you leaving your car at 100% there for any time!!
Do you not have a granny charger or access to a 7kW charger at home?

 

[Jomarkh]

I have the 72 plate mg5 long range. Would putting the car on a Tesco 7kwh type 2 for couple of hours would this balance the battery if taken to 100% from about 80%

If the battery cells are out of balance due to Rapid charging then it may take 2 hours just to balance so might not get done in time.

 

[Cocijo]

It can of course, be done in stages if one attempt does not fully complete the balance.

 

[Archev]

It can of course, be done in stages if one attempt does not fully complete the balance.

Agreed - my Rolec charger is set to charge fully by 09:00 each day. It is also set to only charge during the three off peak periods I have each day. So it will restart an equalising charge (if required) when it comes on again at the next off peak time.

 

[Tigger]

I've just charged mine to 100% and left it on the charger after it reached 100% It kept drawing just over 4kW for about half an hour after it had reached 100%, which seemed quite a lot!

I have just had the same experience. I set my Trophy to charge from 80-100% overnight, and left it to equalise. I expect a 20% charge like this to cost me 13-14kwh, but last night my Pod Point says I drew 18.4kwh, an excess of about 5kwh.

I doubt all this can be attributed to equalisation. Probably what happened is that my % charge figure had drifted off, so although it said 80% to start with, in fact it was more like 78%. So having charged up 20%, in 2 hours, it was only at 98%, so had to carry on for another 20 minutes to get to 100% - only then starting equalisation.

After 90 minutes or so of equalisation, the car was taking 0.04-0.1kw (40-100 watts), so presumably it was mostly done by this stage.

 

[Ayoull]

I have just had the same experience. I set my Trophy to charge from 80-100% overnight, and left it to equalise. I expect a 20% charge like this to cost me 13-14kwh, but last night my Pod Point says I drew 18.4kwh, an excess of about 5kwh.

I doubt all this can be attributed to equalisation. Probably what happened is that my % charge figure had drifted off, so although it said 80% to start with, in fact it was more like 78%. So having charged up 20%, in 2 hours, it was only at 98%, so had to carry on for another 20 minutes to get to 100% - only then starting equalisation.

After 90 minutes or so of equalisation, the car was taking 0.04-0.1kw (40-100 watts), so presumably it was mostly done by this stage.

It will be balancing + loses as the podpont reports what it's passed to the car not what has been input to the battery, and the BMS in MGs uses open circuit voltage levels to determine charge, as such until all cells were at the 100% level and the BMS detected a peak above the maximum in float voltage it will continue charging.

So you have had an imbalance requiring 5kWh to correct inclusive of loses.

The BMS shows 100% once the terminal voltage is at the predetermined pack maximum voltage at which point balancing commences. It's not possible for the BMS to report 100% erroneously as it is terminal voltage based.

Above is the graph from our MG ZS EV MK1 charging and balancing it took just shy of 2kWh to balance the battery and it is balanced evey charge, from empty the MK1 takes around 4hrs and 3kWh to balance following the BMS update.

 

[philip456]

If I stick my MG ZS LR on a PodPoint overnight, will it

(1) Stop charging when it gets to 100% or continue to do a balancing charge?
(2) If it does a balancing charge, will it still be using electricity and charging me more money, while it does the balance?

 

[Ayoull]

If I stick my MG ZS LR on a PodPoint overnight, will it

(1) Stop charging when it gets to 100% or continue to do a balancing charge?
(2) If it does a balancing charge, will it still be using electricity and charging me more money, while it does the balance?

Yes it will use electricity while balancing anything between 0.5kWh and 3kWh depending on how far out of balance the battery cells are.

You can chose not to ever balance but your battery will degrade far faster as you will be stressing the battery pack as the voltages will gradually get further apart between cells resulting in premature cell wear.

 

[Tigger]

Does the battery have to be charged to 100% to balance?

a couple of days ago I charged my Trophy to 80% overnight with the granny charger which achieved the required level by 8am. However, it then charged for a further 1½ hours (3.3kWh) after which time the charge rate dropped to 0.1kW for a period before iSMART showed charging had stopped.

So does that mean the battery is balanced or do I still need to do a 100% charge?

To which Sloucher responded that it had to go to 100% to equalise.

Maybe so, but there is something odd going on here, with the 80% charge drawing extra power.

I have just done this test, starting charging my Trophy at 78%. It charged at 6.2kw for 14 minutes before it reached 80%, which is what I would expect. Then it carried on charging at 6.2kw for another 4 minutes, dropping to 4.6kw for 2 minutes. Next it entered a variable state, charging between 0.06 & 3.93kw and back again repeatedly, for another 4 minutes. Finally it stabilised at 0.04kw or thereabouts for 6 further minutes, before terminating.

This looks to me like equalising behaviour. The gross power leaving the Pod Point charger was 3.2kwh. If we assume 92% efficiency, this equates to 2.94kwh actually reaching the battery. I estimate that increasing the charge level from 78%-80% took 1.3kwh or so, leaving 1.6kwh unaccounted for. If this wasn't used for balancing, what was it used for?

 

[tsedge]

To which Sloucher responded that it had to go to 100% to equalise.

Maybe so, but there is something odd going on here, with the 80% charge drawing extra power.

I have just done this test, starting charging my Trophy at 78%. It charged at 6.2kw for 14 minutes before it reached 80%, which is what I would expect. Then it carried on charging at 6.2kw for another 4 minutes, dropping to 4.6kw for 2 minutes. Next it entered a variable state, charging between 0.06 & 3.93kw and back again repeatedly, for another 4 minutes. Finally it stabilised at 0.04kw or thereabouts for 6 further minutes, before terminating.

This looks to me like equalising behaviour. The gross power leaving the Pod Point charger was 3.2kwh. If we assume 92% efficiency, this equates to 2.94kwh actually reaching the battery. I estimate that increasing the charge level from 78%-80% took 1.3kwh or so, leaving 1.6kwh unaccounted for. If this wasn't used for balancing, what was it used for?

Possibly it is just taking time to home in on exactly 80%?

 

[EV convert]

When I charged mine, my initial thought was that it was battery equalisation and with a sophisticated charger there is no reason why it couldn't do an equalisation charge at any required state of charge.

I guess the thing to do is to charge to 80% and let it do its 'balancing' thing. Once it has finished charge immediately to 100% and see how long the equalisation takes when fully charged and how much energy is consumed during that period.

If it takes, say, 5 minutes instead of 15 and 0.25kWh instead of 1.6, then that could answer the question. If it looks like a duck, walks like a duck and sounds like a duck...

This could be good news as people who regularly use AC chargers to 80% wouldn't need to go the the full 100% which would help maximise (possibly extend) battery life.

 

[Ayoull]

When I charged mine, my initial thought was that it was battery equalisation and with a sophisticated charger there is no reason why it couldn't do an equalisation charge at any required state of charge.

I guess the thing to do is to charge to 80% and let it do its 'balancing' thing. Once it has finished charge immediately to 100% and see how long the equalisation takes when fully charged and how much energy is consumed during that period.

If it takes, say, 5 minutes instead of 15 and 0.25kWh instead of 1.6, then that could answer the question. If it looks like a duck, walks like a duck and sounds like a duck...

This could be good news as people who regularly use AC chargers to 80% wouldn't need to go the the full 100% which would help maximise (possibly extend) battery life.

The balancing is controlled by the cars BMS, as there are numerous ways of achieving balancing.

There's active methods utilising bypass FETs, using capacitors, and using cell module isolation.

Then there's passive methods basically just waiting until the voltage reaches a set level and stopping the charge, and hoping overtime the cells sort themselves out.

EVs all utilise active cell balancing as the passive method doesn't lend itself to extended cell longevity.

All the active methods above are controlled by the BMS usually at voltage determined levels, and are statically fixed to prevent software problems resulting in no balancing taking place.

 

[Cocijo]

The balancing is controlled by the cars BMS, as there are numerous ways of achieving balancing.

There's active methods utilising bypass FETs, using capacitors, and using cell module isolation.

Then there's passive methods basically just waiting until the voltage reaches a set level and stopping the charge, and hoping overtime the cells sort themselves out.

EVs all utilise active cell balancing as the passive method doesn't lend itself to extended cell longevity.

All the active methods above are controlled by the BMS usually at voltage determined levels, and are statically fixed to prevent software problems resulting in no balancing taking place.

Can a BMS equalise a battery pack at any % of charge and be effective? Most will say the pack equalises after a full charge which, depending on the type of cell, should be carried out at least once a month. It would seem to make sense to equalise at say 75-80% of charge rather than having to fully charge, if this is possible and effective? Or indeed why not at every ac charge?
Or am I talking twaddle?

 

[siteguru]

It is possible to do that. But (AFAIK) that's not how any BMS is designed. I believe it is easier to balance at full capacity as cell differences are more noticeable ... the BMS knows what the reference value should be for that SoC so it equalises all cells to that value.

Trying to do that at a lower SoC would likely require far more sophisticated algorithms, as it would have to take account of differences in charge rate between cells.

#speculation

 

[Cocijo]

I suppose that a 0.1v cell difference at 100% charge might be much less and harder to detect at a lower state of charge?

 

[Ayoull]

Can a BMS equalise a battery pack at any % of charge and be effective? Most will say the pack equalises after a full charge which, depending on the type of cell, should be carried out at least once a month. It would seem to make sense to equalise at say 75-80% of charge rather than having to fully charge, if this is possible and effective? Or indeed why not at every ac charge?
Or am I talking twaddle?

In theory yes, but would add a level of complexity that you wouldn't want in the BMS another alternative would be a two point balance system, where it could balance at 80 or 100% as predefined levels.

As far as I know no BMS implement this as the point of balancing is to prevent voltage strain in particular overvoltage which isn't a concern at 80%

If the pack has not been used for DC charging the state of imbalance would be low, however on the MK1 ZS EV, after a BMS update it balances on every charge now, regardless of how it's been charged previously.

 

[EV convert]

I bow to your greater knowledge however, I see no harm in testing the 80% + 'balance' procedure followed the full 100%. If it appears to work I will look to extend the 100% charging intervals and keep a close eye on how long the balancing phases take as well as how much energy they consume.

 

[Tigger]

I see no harm in testing the 80% + 'balance' procedure followed by the full 100%.

I like your test - go for it! It occurs to me that one reason we have been advised to balance at 100% is that this reliably recalibrates the % charge estimate. At 80% it cannot be recalibrated, and remains an estimate. (This doesn't bother me.)

 

[Ayoull]

At 80% it cannot be recalibrated, and remains an estimate. (This doesn't bother me.)

On the NMC packs the BMS will only be -/+1-2% out on SOC calculations, due to the nature of the voltage drop across utilisation of the pack,

Compared to LFP where it will be around -/+10% out hence the need for weekly balancing, only at the extremes of the SOC will the LFP BMS accurately read out as the voltage drop is similar to that of NMC, but plateaus between 5 and 95% charge as the voltage remains stable between these points.

 

[EV convert]

Thanks folks,
This seems to start making sense now. I am having a Zappi fitted on Monday and will try out the 80% and 100% charging tests very soon and report back (if nobody's beaten me to it!).