HOW-TO: Recovering lipo packs after a short.

[neptronix]

Well, it's quite possible that this high altitude is making me stupid. There has been a recent rash of dumbass maneuvers on my end.

Recently i wired up an cell_man 9FET infineon controller's power leads backwards. A puff of nasty smoke came out of the 10S 5AH 20C turnigy lipo pack i hooked it up to, surprisingly, the controller lived..!

The connectors were quite charred afterwards.. after cleaning them up, i noticed that only one 5s lipo brick was smelly.. and seemed to have 1 cell WAY off..



So i plugged it into my 2x 50w halogen array to see if it would transfer current.. nope. The cell at 0.300V would drop to 0V in an instant. I figured.. i had burned a connection in the lipo.. so i had to investigate.



There's the dark spot, looks like cell #5 got hit.



More gore after taking the tape off..



WOW.. something fried in there didnt it..



The padded tape that protects the top of the pack had solder on it and scorch marks. Looks like the solder leaped off the circuit board!



LOL.. yep, that's exactly what happened. It cut the connection from cell 4 to cell 5 almost completely.
Now i'm wondering if the solder is designed to do this, in order to prevent even worse shorts that lead to a runaway thermal reaction inside.

Perhaps this pack is not toast.



3.98v on this cell looks good to me. Let's rejoin the cell bridge and see what happens.



First, tape off the adjacent cells adequately for the love of god. You don't want to cross-solder any of the cells and have the solder explode / vaporize in your face, killing a cell in the process!

Another thing to note when soldering here, is that the solder used on this pack is a bit different than the standard lead/rosin core stuff. You can definitely blend standard solder in with it, but you can't really solder to the aluminum (???) tabs so well.. you'll have to mix up the lipo pack's solder with yours.. you'll know it's mixed as the solder used on these has a very different smell when heated up..



Bam, there it is.. 10 ga. copper wire soldered between the pit of despair, doing it's job as a bridge ..
( 10 gauge is fine here since this pack is used in series, if you're doing this same repair to a turnigy nanotech, 45C lipo etc though, you might want to beef it up and go for 8 ga.. )



The CellLog 8S approves of my repair.



..and so does the iCharger! the IR on the bridged cell is actually reading at a lower resistance level than the other cells.
Note: the iCharger is awful at measuring milliohms, it will provide a random variance of +/- 5mOhm, which is why these numbers are all over the board. But it does a good enough job of determining if there are severe anomalies.

A discharge test at 3.85A confirms that the cell

Next: i will do a full discharge test of this pack. But i am quite confident going off the IR that this cell has not received much damage, if any. It hasn't puffed or change voltage.. these are good signs.

If you have a pack that has been shorted, and hasn't puffed.. you may want to inspect it for this exact kind of damage. This repair is not difficult.. definitely one of the easiest lipo surgeries.

Hope this helps

[jonescg]

Interesting that for the odd-numbered packs they still want to put the leads on the same side, so they solder that diagonal bridge in. Being copper, the tabs solder up nicely don't they

[Philistine]

I have done that to many packs, and after resoldering the tab, all packs have gone back to perfect performance.

[neptronix]

I have done that to many packs, and after resoldering the tab, all packs have gone back to perfect performance.

Really? every time!?
Does the same thing occur - the link between cells, namely the last one towards the positive lead, just totally blow like mine did?

Or in your instance, does the tab itself unsolder itself?

I could not get any solder on the tab whatsoever, even using a good application of flux. i think it requires some special solder that i do not have... that's why i mixed up my solder with what was already there..

RC Lipos are intense, aren't they? they'll just vaporize / explode anything in their path. Little 5AH gods of thunder.

[Philistine]

Yeah, mine have always blown the linking solder like yours. When I first started using lipo I used to break down and reconfigure packs with more AH/ or change voltages, but I ended up making myself choose semi-permament battery configs, as I couldn't stand constantly KFFing myself. It is amazing how sometimes you can see yourself doing it before it even happens, but you seem powerless to avert the idiocy.

[neptronix]

Yeah, mine have always blown the linking solder like yours. When I first started using lipo I used to break down and reconfigure packs with more AH/ or change voltages, but I ended up making myself choose semi-permament battery configs, as I couldn't stand constantly KFFing myself. It is amazing how sometimes you can see yourself doing it before it even happens, but you seem powerless to avert the idiocy.

A ha.. i'm surprised nobody mentioned that this was the common death route. I'd read a few threads about people shorting their lipos and then just giving up on em.. so i figured i'd post up..

I also reconfigure my lipos almost on a weekly basis. I think i just need to get some motor winds that all run on 72v or so and permanently fix things together as you have. Time to stop screwing around with all the wiring constantly.

[miuan]

I also reconfigure my lipos almost on a weekly basis. I think i just need to get some motor winds that all run on 72v or so and permanently fix things together as you have. Time to stop screwing around with all the wiring constantly.

That's what I've done!!!
But even so, I can't avoid a spark now and then.
Luckily I use thin wires, so I never get caught killing my battery.
But I pay the duty by 5V sag at full power!

[dogman]

Interesting that it didn't vaporize the tab, while hot enough to melt solder.

All my boo boos resulted in the tab itself vaporising, Not the soldering tab, but the meltdown was where the foil tab exits the cell. So it's unrepairable. I desolder or slice out the bad cell, usually it's on the outside, and move the red wire to a new location, creating a 4s pack out of the toasted 5s pack.

[wineboyrider]

A ha.. i'm surprised nobody mentioned that this was the common death route. I'd read a few threads about people shorting their lipos and then just giving up on em.. so i figured i'd post up..

I haven't given up on mine, yet. Actually I got a running 11s lipo pack now. My cells died in my KFF event. The zippy 8ah cell just died when I was driving to work. I tested the cells and the tabs the one cell still has 0v? I still think it's a dead cell, but definitely worth another stab with the multimeter.

[Ykick]

viewtopic.php?f=14&t=27005&hilit=+Fuse

[amberwolf]

I could not get any solder on the tab whatsoever, even using a good application of flux. i think it requires some special solder that i do not have... that's why i mixed up my solder with what was already there..

With the Turnigy packs I got from Mdd0127, both seemed to have normal solder and AFAICT copper tabs. Each cell in both packs had a copper tab sort-of compression-welded to the aluminum tab, and the other tab appeared to already be copper (or at least, a copper alloy that is easy to solder to).


I do applaud the idea of the tape over the adjacent pads/tabs; I probably should have done that on mine but I didn't have any problems, thankfully.

(On another high-power experimental pack with thick copper busbars, I once dropped a tiny lockwasher down between teh bars, and it vaporized instantly, thankfully not leaving anything behind to continue the short. I have a feeling even the little screw that washer was for would ahve done the same thing, though ).

[dnmun]

amazing that the controller did not blow the caps and shunt. usually that is fatal for the controller.

[amberwolf]

True. The controller that has been running CrazyBike2 since it's resurrection as a hubmotor bike was donated to me damaged like that, but fortunately all it needed was caps and shunt, and has worked fine since then. (The Watts Up that was in series with it wasn't so lucky)

[neptronix]

http://endless-sphere.com/forums/viewtopic.php?f=14&t=27005&hilit=+Fuse

Good infos! Well, it sounds like that side really is a fuse.
Thank you turnigy/zippy/haiyin/herewin/ whoever

Dogman: Maybe the difference for you is that you have your lipos in parallel, then series? or you have something other than these 20C units? that's all i can think of. I run all 20C and go series, then parallel with mine.

Amberwolf: I don't see any copper in mine, just a shiny tab that won't allow any solder to stick to it. Maybe my soldering iron was set too low.. oh well, the IR was lowest on that cell so i'm sure the bridge was adequate.

I started using tape when i accidentally shorted one of the lipos with the soldering iron ( probably cross joined the solder ). I thought i was safe, but a little twitch of the hand from the morning's caffiene jitters and there went a cell

[dnmun]

i think you can repair the controller like amberwolf did with the one he rebuilt. check the input caps with the ohmmeter in your DVM, put the red probe on the large red wire only, not the small red wire, and the black probe on the negative of the controller, and the caps will show up as a changing resistance value if they still work, then reverse the probes briefly and watch the resistance value change back in the opposite direction. you have to put red on red first or you will reverse the polarity of the caps with the ohmmeter too. but that much current out of the cells indicates the caps are shorted or the shunt took all the current and it should be blown too. more likely the caps are open circuit now and the shunt is vaporized.

[parabellum]

4s 5A turnigy hard cases are flipped so no solder link, maybe that why I always end up with vaporized 4mm bullet connectors, once 6mm connector got liquid as well. Maybe must just learn to connect quicker.

[dogman]

Not much metal to those male 4mm bullets. Maybe they just vaporise faster than a tab.

Most of my KFF's have been with paralell paired packs, then get lost in the tangle and connect a pairs + and- together. Zap, and a tab would be fried on the first cell of both packs.

I would guess it just finds the weakest link, tab, solder, or bullet. In one case, I did get lucky and have an anderson melt fast enough to have no damage to the pack. I learned slow, and did it quite a few times.