Acceptable voltage difference in 22S/22P lithium ion pack?

Summary:
22S/22P pack consisting of INR Samsung 25R cells (purchased brand new). Cells are connected with .25mm thick pure copper bus bars. The pack uses a BMS with 150mA balance current. Pack used to pull 500A short burst current and 300-350A constant current.

Series connection #22 has a voltage difference of 98 mV. I will balance the pack but what voltage difference is acceptable for a pack like this. Thanks.

how long is a piece of string?

up to you, so long as LVC / HVC are triggered by cell voltages, it's just a matter of small reduction in pack capacity

Often the delta when balancing stops is hard coded in the BMS

Target for most good balancers is 5-10mV, some even claim 3mV.

within ~10-20mv (.01V) of one another is fine

If adjustable and sessions go too long, I'm OK with 30mV.

5mV is too tight a delta IMO



chronically imbalanced by more than 20-30mV at charge termination, it's probably approaching time to consider pack retirement.

is that the group that powers the bms?

john61ct said:

Target for most good balancers is 5-10mV, some even claim 3mV.

within ~10-20mv (.01V) of one another is fine

If adjustable and sessions go too long, I'm OK with 30mV.

5mV is too tight a delta IMO

chronically imbalanced by more than 20-30mV at charge termination, it's probably approaching time to consider pack retirement.

Click to expand...

Good info. Thank you for the response.

goatman said:

is that the group that powers the bms?

Click to expand...

Good question. I’m checking with the manufacturer to see if the BMS is powered by the last row of cells

22p is a huge pack. I doubt the BMS current draw would be enough to bring the end cell down that much.

I would double check with a voltmeter directly to the cells if possible. Sometimes there will be a voltage drop in the last balance wire due to current draw of the BMS or something downstream that causes a slight error in the reading. Same thing can happen during charging where the end cell "sees" a higher voltage and limits the charge on that cell which results in a lower voltage during idle or discharge.

If you do the math on how much % state of charge that voltage difference represents, I suspect it won't be more than a few %.

fechter said:

22p is a huge pack. I doubt the BMS current draw would be enough to bring the end cell down that much.

I would double check with a voltmeter directly to the cells if possible….

If you do the math on how much % state of charge that voltage difference represents, I suspect it won't be more than a few %.

Click to expand...

Good info and suggestions. Really appreciate it.

If your independent voltmeter measurement still shows that one group low, you could charge it with a single cell charger to try getting closer to the rest and see if it stays there. I remember certain tool packs always having issues with the end cells due to voltage drop in the wiring throwing the BMS off. If this is actually your problem, it may be possible to add a wire somewhere to avoid the issue.

I believe I found what may be the issue. The last cell in row 22 had a “questionable weld“ so I removed it. That particular lithium ion on cell was at about 3.7 V. That is likely culprit since it was probably dragging down the voltage for the entire row of cells.

I removed the 22nd cell in each row so now the pack is 22S/21P instead of 22S/22P. This avoids any problem of adding a brand new cell (with higher capacity) than the other cells. Fortunately it was on the end of the cell row so it was easier to remove. I will let you know how it turns out next week.