locating bad elecricaly leaky cells in a li-on battery pack

[ME]

I have a 13 10p pack with a bt bms ,There are 2 or 3 cell groups that loose about 100/200 millivolts overnight , I guess the groups that self discharge have a bad/leaky cell in them , Is there anyway to find the faulty cells without stripping the pack down and testing each cell from the groups that self discharge?, The cells are samung 18650 15q lithium colbalt types. eg one group was at 4.2v last night , this morning it was at 4.135v, that is by looking at the blutooth jbd app.

[Siwastaja]

Leaky cell or leaky BMS. One interesting test is to disconnect BMS instead and see if the problem persists.

If it's a cell, I see no way except physical disconnection of the paralleled cells, unfortunately.

Also note it's not necessarily leakage current, could be just imbalance; do they seem to hit 4.2V top voltage at the same time as others?

If you keep it longer than overnight, e.g. a second night after taking that reading, does it continue to drop? If so then it's leakage current yes.

[TERRA Operative]

Would a thermal camera show which cell is self discharging?

[Siwastaja]

Maybe if it can show differences of 0.001 degC...

[DavidAlberto]

To successfully visualize these leaks, you need a high-quality thermal core. In lower-end cameras, the NETD will likely drown out the subtle heat signature of a 'leaky' cell, making it impossible to distinguish the actual fault from the background noise of the image.

It would be wise to first perform the check suggested by Siwastaja to confirm that an electrical leak is indeed the root of the problem you're facing.