A failure in a 12V Lifepo4 battery

[r111]

I have had this U1-12RT 40AH LiFePo4 Valence battery for few years. Use with a trolling motor on my kayak.
The battery is very robust, nice BMS, used widely in medical field etc, and is readily available as used item, for as low as $70, may be even free if you can score these at local recycler, hospital etc.

For few years I enjoyed full 40AH capacity - which I measure carefully both during use and charge cycles.
And then one day, it drop very suddenly to just about 10AH, both in use and charge.

I opened the case, BMS doesnt have any signs of burned out elements. When charging one of the 4 #P packs directly (there is a balancing connection), I see that it, unlike its other 3 in-series siblings, takes forever to take charge. No signs of trouble - no smoke, heat, short circuit etc.

What kind of failure mode could it be ? I am thinking about opening the case and getting to the individual cells, to try to identify and cutout/replace the failed cell. Since I know the bad #P pack, it should make it easier to ID the bad cell , I hope.

Anyway, all and any advice is welcome. If the battery slowly lost its capacity, I'd may be let it go. But this was so perfect @ 40AH for few years, never showing any signs of deterioration, I hope I can save it ?

 

[LeftieBiker]

Can you see the cells at all, to tell if one or more of them are swollen?

 

[r111]

Havent opened the case yet and only have a vague idea how to. But this is my guess as well - that one of the cells is bad and is draining its peers in the parallel pack that it is part of.

There is a Youtube of some1 opening 100AH version of same battery, I hope mine opens the same way.

The entire battery is fine mechanically, havent been dropped or pierced or anything like that.Dont see any signs of bulging either.

If one of the cells is bulging, I better open the pack soon, before it deforms and ruins its neighbors ?

 

[r111]

She gets a sound 1 on scale of 1-10, in terms of ease of disassembly, 1 being damn near impossible.
Seems the top and the bottom plastic parts are almost thermally welded somehow. I used a heat gun,
but could not get them to separate, so I ended up chiseling away the overlapping part.
Will do a video at some point.

120 of 4.4Whr (on the label) Valence lifepo4 cells are used, in 30p4s configuration. Quick math confirms the ratings (each 30p pack has 40AH capacity, for 40/30 = 1.333 AH per cell * 3.3V, derated)

Welded cells, with some clever folding of copper plates.

Bottom copper plate has signs of corrosion. The affected 30p pack is welded to the bottom plate. I think some salt water (ouch !!!), may be 20ml or so, got in during one of my kayak outings, ended up on the bottom of battery - since I store these upright.

I think the corrosion residue is shunting (not 0 Ohm short, but through whatever resistance is there) some of the cells in that pack, so it is very slow taking charge and looses the charge quickly . Didnt see any swollen cells, at least on the easily observable exterior part of the pack

Some cells have lost the exterior paint, so the corrosion residue is touching from bottom plate to the battery's exterior shell, which is likely connected to the other terminal of batteries.

Another 30p pack is welded to the same corroded bottom plate, but since it is welded to the other terminal, which is already connected to the battery exterior, that pack is not being shunted by corrosion.

Thats my theory for now.

Will need to figure out a way to get rid of that corrosion somehow

 

[EbikeAus]

Aren't these 18650 cells inside?
Shouldn't be any swelling but salt water sounds bad.
Have you seen DIY tech and repairs on YouTube he has tonnes of the valence packs modded with batrium bms.

The flat folded copper bus is common in these packs.
I'd be making sure you remove all corroded cells