Electronic – 18650 battery pack with “dead” cells, other cells still usable

bmsliionlithium ionover-dischargeover-voltage-protection

I have an 18650 battery pack (12p20s so 240cells) with 2 parallel groups which recently "died" (voltage is 0v or very close to 0v (0.09v for one of the packs 0.0v for other parallel group). All other parallel groups in the pack are approximately 3.6-3.7V. The (dead) cells most likely died because I discharged/charged the cells without using a BMS for a long time (since my bms had broken and I was too ignorant to replace it right away).

I am quite sure the cells were never over-discharged (<2.5v) since the total pack voltage remained above 65v (3.25 average). However, I am sure the cells were badly overcharged for many charge cycles (seeing as when I decided to replace my BMS with a working variant eventually, some of the cells were already 4.2V when the total pack voltage was approx 78v. So I'm quite sure that killed them…

My question now is: Should I even consider re-using cells from this pack for building a new battery pack? Or should I replace all cells?

Does testing the cells for capacity and internal resistance give a good indication of how much the cell was overcharged/overdischarged/overheated? Or is there no real way to measure the 'health' of these cells properly? How much of a negative effect does overcharging and discharging have on capacity and internal resistance?

Thank you very much for any input, it is greatly appreciated

Best Answer

Are my "good" cells good?:

It is likely that the cells that test at 3.6-3.7V are entirely acceptable for future use. If a pack becomes severely unbalanced the more discharged cells get driven into deep discharge while the other cells are still within the normal operating range. If the discharged cell internal protection operates you have a mix of cells, some of which have never been subject to marginal conditions.

Cycling the "good" cells in a tester to confirm their capacity will give a reasonable indication of their condition.

You did not directly ask about the 0V cells but they too MAY be in better condition than you expect. And may not ...

Are my "dead" cells dead?:

Recovered cells may "never be quite the same ever again" so you'd be advised to characterise and test them if they were able to be recovered.

It is POSSIBLE that the 0V cells have had internal protection operate to disconnect the cell. It MAY be saveable. ie the actual cell MAY still be above the really-dead level). Try charging from a 3 to 4V source via a resistor - say 100 Ohms or more. IF Vbat reappears it may still be OK.

Would I have to manually reset the protection of the 0volt cells before trying that (charging via resistor) in order for that to work? Wouldn't that be dangerous since there must have been a reason the protection was triggered?(protection was not triggered by a too high discharge current since I'm 99% sure no cells were externally shorted and my load only draws 5amp max (per cell) whilst this model of 18650 should be able to handle 10A (per cell)? Thank you

It all deep-ends. If the cell is drained at a rate below where current overload protection cuts in it will reach a lower voltage limit. This is probably around 2.5V. The internal protection disconnects the cell from the external load. The cell is still subject to self discharge (slow) and some monitoring current (also small) and will SLOWLY head towards the about 2V where it may be beyond recovery. With luck the period between 2.5V and 2V can be very long.

The internal protection circuit may well allow trickle up from that state. Depends on manufacturer and will (very probably) do no harm at very low currents. The capacity at this voltage level is so low that charge up is quite quick. eg (figures out of my head) as say 3000 mAh cell may have 1% - 5% capacity there. At 1% that's 30 mAh so 30 mA should bring it back in an hour (but maybe more) and it may be "lurking very close to the boundary and come back in minutes. I'd "hope" that say <= 100 mA would do no harm (300 mW abs max at 3V and possibly only 50-100 mW)

BUT do this somewhere "safe".

Related Topic