Question on LiPo Cell Failure Mechanisms

[GMUseless]

So I've have a couple of HK Turnigy Lipo 6s 3Ah packs fail. But the method of failure has me guessing as to what has happened.

In one case, the pack arrived with a missing cell (ERR on the CellLog, 19.25 volts on a 6S pack)...I sent it back to HK.

In the second case, the pack tested fine. I cycled it from 80% to 40% three times before going into service. These tests results were all good. Then, after 20 or so application cycles (90% down to 25%), I pulled a few packs off and stored them at 80% charge for about two months.

When I tested them this month, one pack showed a ERR on the CellLog. The 3rd cell from the negative end was missing. Almost like it is shorted through. All of the other cells show 3.95v, and the pack total is 19.75. This appears to be the same issue as the first bad pack I returned.

I can understand that a tab can become unsoldered or loose...but if this happens, would you not lose the voltage from the main leads. The cells are after all in series: if one connection drops, the entire pack drops. Am I missing something here?

I cut the casing and tape from the pack, and all of the soldered connections look good.I tested each solder point with a volt meter, and indeed all cells show 3.95 volts, except for the one that shows 0v. Is this cell shorted?

[neptronix]

That's curious...!

FYI.. are these turnigy lifepo4? because those cells like to fail like that - sudden 0v.

Otherwise, this is a very strange failure mode.

[GMUseless]

Yeah Nep, it's definately LiPo (hence the 19.5 volts for 5 cells.) It's almost like an internal shunt has formed. I understand this can happen with the Li-Ion cells using carbonate ester electrolytes...maybe there is a simlar mechanism in the polymer.

[amberwolf]

I dunno what the exact mechanism is, but it is a very common way for them to fail, apparently. Lots of people get packs even brand new like that, with 0V cells (I have some here), and others just die later like that. I'm sure occasionally it's a broken balance wire, but more often the entire pack is down the voltage of that cell, showing that the cell itself is essentially dead but conducting.

I would not want to discharge or charge the pack without removing the dead cell first, so it's not having current pass thru it, but so far the ones I've revived by doing that are working for testing. Not yet tested on the bike or with heavy loads or fast charging.

[texaspyro]

And remember that all packs (regardless of chemistry) have a built in pain-in-the-ass detector. They will fail (in the most spectacular way possible) at the exact time that will cause you the most grief possible.

[Ypedal]

I would absolutely not.. repeat NOT discharge this pack....

OUTSIDE.......

Tap into that single cell via the ballance leads and apply a small charge.. like 1 amp.. see if the voltage goes up , once you make it to 3v remove the charger and see if it keeps the voltage or if it drops right away back downt 0v.

report back.

have camera handy in case of large fireball... we like pictures !

[Kin]

Well, I guess the barrier separating oxidizer and reducer sides could have broken down. One of the reason why overcharge is bad is that it damages this membrane that keeps the battery not shorting itself. But it's very weird that it would die spontaneously, so maybe the suggestion I offer is bad. sorry for your loss.

[liveforphysics]

Impurities or tears or pinholes in the separator can do that.

I've watched the ladies assemble them. Maybe one had a sharp nick in her fingernail and it keeps making pinholes.

[Kin]

Impurities or tears or pinholes in the separator can do that.

I've watched the ladies assemble them. Maybe one had a sharp nick in her fingernail and it keeps making pinholes.

With all respect to the assembly workers, SHE MUST BE TRACKED DOWN.And...[her fingernail]... eliminated.


Edit: In retrospect I don't know if I can ever say 'in all respect to assembly workers.' Ethical struggles. There is a huge disconnect between realities here.

[docnjoj]

Hmmmmm...............Sort of like the problems in some condom factories.
otherDoc

[GMUseless]

Thanks for all the feedback. I definately won't be charging any suspect cells. It might be cool to see what happens...but for $20, I'd just assume trash it an buy a new brick. I've got LiFePo4 cells for misc. tasks...so I don't really have a need for recovered LiPo that may or may not be prone to further failure.

After some research, I've found that in Li-Ion cells with liquid electrolytes, the separator is designed to melt during a thermal runaway...effectively nullifying it's pore structure and shutting the battery down. Does anyone know if the separators in polymer electrolyte batteries are designed to perform the same shutdown?

If so, then I could buy that as a mechanism for failure: something finally gave way internally, the battery shorted and separator served it's shutdown function. That would explain the 0v / yet seemingly conductive outcome as well.

Though a bit dated, some good info on battery separators was found here: http://tinhoahoc.com/Battery/cr020738u.pdf