Voltage difference between LFP batteries connected in series

[Awsmits]

I have two EVE 280 Ah LFP batteries (12V each) connected in series (24V) each with its own BMS. The BMS settings are identical. They are charged through a Growatt 3000W inverter through either solar or line AC. Early on the individual cell and total battery voltages were very close, but lately their normal charged status are quite different, say 3.35 vs 3.45 (cell averages) and 13.4 vs 13.8 (battery average). I’ve set the Growatt to charge to 27.5V, but that doesn’t even out the V differences between the two batteries. Am I being too concerned about this difference? Is that difference normal?

 

[sunshine_eggo]

I have two EVE 280 Ah LFP batteries (12V each) connected in series (24V) each with its own BMS. The BMS settings are identical. They are charged through a Growatt 3000W inverter through either solar or line AC. Early on the individual cell and total battery voltages were very close, but lately their normal charged status are quite different, say 3.35 vs 3.45 (cell averages) and 13.4 vs 13.8 (battery average). I’ve set the Growatt to charge to 27.5V, but that doesn’t even out the V differences between the two batteries. Am I being too concerned about this difference?

Not concerned enough.

Is that difference normal?

Normal in that it happens, but it's not good and steps should be taken to make sure it doesn't happen.

When you use 12V LFP to make 24V or 48V, you have to periodically break the bank down and fully charge each as 12V. Battleborn actually mentions this in their battery manual.

For a DIY made of cheap gray market cells, it's likely even more important.

If I were you, I would get an 8S BMS, top balance the 8 cells and build a single 8S battery.

 

[curiouscarbon]

this device is one option that might mitigate the need to break the packs apart occasionally

Battery Balancer - Victron Energy

Equalize the state of charge of two series connected 12V batteries using the Battery Balancer. Find a Victron Energy dealer near you.

www.victronenergy.com

good luck!

 

[curiouscarbon]

If I were you, I would get an 8S BMS, top balance the 8 cells and build a single 8S battery.

if switching out the two 4S BMS with one 8S BMS is an option, i second that.

 

[sunshine_eggo]

IMHO, while I let Victron have their way with me regularly, I would never buy one of their balancers or their batteries. The former due to both their cost and their anemic current.

Kilovault and many others make 24V balancers that seek to maintain both 12V at the same voltage and actually transfer charge from high to low. They also do it at higher current.

 

[Awsmits]

Ok, thanks for the input. Question is, can I leave my 4S BMS units in place to balance the cells within each battery, while connecting a Victron or Kilovault balancer to equalize the voltage between batteries? I hate to think I bought the 4S BMS units for nothing.

 

[toms]

Very common (inevitable actually) to see this happen. Without disassembling anything you can get a 12V charger and when your system is fully charged you individually top up all batteries to the same level.

 

[Goboatingnow]

My view is in a series string , what you are seeing is simply cell out of balance , clearly your charger isn’t carrying out any balancing. Out of balance merely means some , typically minor loss of overall capacity , but other then that it’s not a safety issue in any way, assuming you are doing cell monitoring

You either need to add cell balancers , or just live with it , the obsession with matching cells in a series string is just that an obsession. Ultimately your capacity calulations should include headroom for out of balance cells.

What I have done is every so often ( 1,2 a year ) is break down the batteries and balance charge each cell separately

My own view is unless yiu have serious deviations in cell capacity ( +20%) , cell imbalance should not be focused on and the battery should have enough capacity headroom to withstand out of balance loss of energy ability

 

[robby]

Unless you have a lot of time to keep balancing the cells whenever they go out of whack I would just fix the problem properly with an 8S bms.

 

[Goboatingnow]

Unless you have a lot of time to keep balancing the cells whenever they go out of whack I would just fix the problem properly with an 8S bms.

The reality unless they are very badly out , there’s no need to balance unless you are trying to squeeze the last drop of capacity out of the bank

 

[curiouscarbon]

switching gears a bit.

if it is only one or two cells significantly high,
manually bleeding off energy by manually connecting a resistor to the highest cell for 5 seconds at a time or so, can provide "fast manual passive balancing" and cheaply temporarily fix.

say you have 3.3v, 3.3v, 3.3v, 3.4v on one pack.

get a 50-100Watt rated resistor, around 1 ohm.

watts = (volts*volts) / ohms

3.65v is max. so 3.65*3.65 = 13.3
13.3 / 1 ohm = 13.3 watts discharge rate with 1 ohm resistor.

example resistor for ~10 usd : https://www.amazon.com/AISIBO-Aluminum-Wirewound-Resistors-Equipment/dp/B07HC8CH2N/

touching the two respective opposite sides of this 1ohm 50watt resistor to the two respective terminals of the highest voltage cell will discharge it at about 10 watts, which is manageable.

edit: the idea is to monitor the voltages from the bms or such, and manually connect the resistor to the highest voltage cell for a few seconds at a time. once the voltages are even, stop.

hope this info is somewhat useful

 

[curiouscarbon]

Unless you have a lot of time to keep balancing the cells whenever they go out of whack I would just fix the problem properly with an 8S bms.

replacing the two 4S BMS with a single 8S BMS would reduce long term issues. it would balance all cells in a group.

---

ok but given that there's already two 4S BMS in service, what do?

to balance the pack, manual resistor balancing might be the least expensive option. if the cells went out of balance once, occasional balancing would likely be necessary.

1) 8S BMS
2) manual resistor balance

two options that might help!

 

[curiouscarbon]

Ok, thanks for the input. Question is, can I leave my 4S BMS units in place to balance the cells within each battery, while connecting a Victron or Kilovault balancer to equalize the voltage between batteries? I hate to think I bought the 4S BMS units for nothing.

(yes)

if the two 4S BMS are otherwise working fine,

1) manual resistor drain highest voltage cell
2) attach bank balancer and monitor for imbalance

two least expensive options imo.

the cost of an 8S BMS might exceed a bank balancer. in this case, with cost as a constraint, manually applying a resistor to balance, or adding a bank balancer, ought to function.

so many ways to the same destination!

thanks for posting

 

[Awsmits]

I’m not sure how the concern for out of balance cells came up. The cells within my two batteries are very well balanced, it’s the voltage difference between my batteries that concerns me. Switching to an 8S BMS sounds like a good idea. I don’t think anyone responded to my question about adding a battery balancer with the 4S BMS in place. Is that Ok? I’m also going to check all the connections within and between the batteries, as I know that can affect current balance.

 

[Awsmits]

Sorry, curiouscarbon did address my question, thanks. The other consideration in play here is that my batteries are in a boat, hard to access frequently, so I’d like a solution that didn’t require frequent access and manipulation. Sounds like a Kilovault 24V ($39) would do the trick, yes? Much less than a good Bluetooth 8S BMS.

 

[curiouscarbon]

if the individual cells in each pack are balanced, and it is just the two pack voltages that are uneven, then yes that should work

ive never used kilovault, but sunshine eggo has more experience than myself, so it sounds like a better option than the 1A victron balancer.

good luck!

 

[Goboatingnow]

Again folks there is no real issue with out of balance cells nor in out of balance series batteries.

Only if it’s a serious effect on overall capacity dues it raise concerns.

For example post charging , bleeding of charge into a resistor is a complete waste of time abd all you are doing is throwing away energy stored in that cell.

It’s only in a series string , where one cell hits charge cutoff , well before the others that overall battery capacity is then reduced. But other then that it’s of little consequence

 

[sunshine_eggo]

The reality unless they are very badly out , there’s no need to balance unless you are trying to squeeze the last drop of capacity out of the bank

The issue here is you've likely never experienced substantial capacity loss due to imbalance. Very common occurrence here. One bloke lost nearly 50% of his Battleborn's capacity because he NEVER charged it to a high enough voltage for the BMS to conduct any balancing.

Countless descriptions of users not able to get anywhere near a typical 14.2-14.4V absorption charge due cells out of balance. Even very small SoC imbalances can cause this.

Folks DIY-ing their own batteries finding the as-received cells varying by up to 20% SoC.

 

[Goboatingnow]

The issue here is you've likely never experienced substantial capacity loss due to imbalance. Very common occurrence here. One bloke lost nearly 50% of his Battleborn's capacity because he NEVER charged it to a high enough voltage for the BMS to conduct any balancing.

Countless descriptions of users not able to get anywhere near a typical 14.2-14.4V absorption charge due cells out of balance. Even very small SoC imbalances can cause this.

Folks DIY-ing their own batteries finding the as-received cells varying by up to 20% SoC.

I’m not saying you don’t need to balance , occasionally , you do , once on receipt , and occasionally 1-2 times a year. Fractional C usage LFP tend to generally stay in balance.
The fact 14.2 14.4 setpoint is irrelevant, , lithium’s can be charged to 95% on lower voltages ,
In the battleborn case if you have a cell reaching the cutoff point but exhibiting 50 % soc you ineffect have a failed cell. , and very few passive balancers could handle that level of imbalance anyway

 

[Awsmits]

Glad to hear the general opinion that my issue is “normal” and not harming my batteries. Through this forum and elsewhere I’ve learned two lessons. First, when my batteries are charged above 3.4 v/cell, I’m really working at 95% SOC, the steep part of the charge/discharge curve, so no big deal if cells in one battery are at 3.4 v/cell and the other battery is 3.5 v/cell, they are both effectively charged. As a newbie, I guess I was being anal about having them exactly the same. Second, I had to sort out in my mind what charging limits I set the Growatt to deliver to the batteries, and what charging limits I set in the BMS. I set the BMS to balance when charging above 3.3 v/cell. Initially I set my Growatt limit too low and the batteries were not reaching full capacity. I now have my Growatt set to cutoff at 27.5V, which seems to bring the cell voltages up to 3.45v.