How's this for a battery torture test? :)

[Niconical]

I made "the" mistake.

I put down a light that was turned on then forgot about it.

Oops

It was an incan lamp run by 2 x AW LifePO4 CR123A cells.

After a while, I realised and rushed back to the light. At first I thought the lamp had blown as the light was off, but it soon became clear that the lamp was fine, it was the batteries, they were..... empty!

The first one I tested had a negative voltage. Yes, negative. I did have the correct ends connected for the multimeter, it was a negative voltage. Only slightly though, I think it was about -0.05 and slowly creeping back up to zero.

It got there, then very gradually it started the climb back up the voltage scale. I left it for a while, maybe 20 minutes, just to see what would happen, and by then it had dragged itself up to around 0.6v.

I then put it in the charger and waited.

It charged, and when it was full I did a runtime test. It got 40 minutes in a Romisen RC-N3 before dimming, then I dragged it down some more, just for good measure

I charged it again, and it was fine, back in the Romisen for some more testing.

Now, I'm not saying the battery is fine. It is no doubt in bad shape, but it does work, and it seemed to perform quite well. I would never include it in the rotation to be part of a pair in a 2 cell light, instead I will do what I have done with previous (intentional) torture testers and put red tape around it and only use it for single cell lights.

I am not recommending dragging any battery down to 0v (or lower....), I'm just showing this as an example. LifePO4. He may not be the biggest kid on the block, or have the most stamina, but he's definately the toughest!

 

[geek4christ]

Cool stuff, Niconical.

Do you by any chance know if there was immediate capacity loss from the polarity reversal? For instance, could you do (or have you already done) a runtime test on the RC-N3 with a known good LiFeP04 cell?

 

[Niconical]

It seemed about the same. When I previously did proper testing of LiFePO4 with that light, I got to 30 mins of accurate timing, then it dims "sometime after that". This tortured cell was the same. Maybe the previous tests were 42 mins, and all I got was 38 mins because of the rough treatment, or something like that, but I tend not to bother looking carefully after the half hour point. All I can therefore confirm is I got a solid half hour of full power use from a cell that had been drained to zero and beyond, and then it charged back up again and did the same thing and is now back in the single cell light rotation

 

[AW]

A negative voltage reading means the cell has been reversed charged by another cell due to -ve balance. The -0.05V is about the max. the cell can take without being killed. I have tested cells to -0.025V and they all seems to rebound and work fine without much adverse effect or capacity loss.

AW

 

[Niconical]

It is a credit to the strength of these AW batteries that I have been unable to actually destroy one. I have red tape on 3 of them, reminding me that those ones were treated badly, and black tape on 2 of them to indicate that they were treated badly enough that by any standards, they should be dead.

I need the tape reminder because without it, just in general use, I can't tell any difference between these cells and the others

 

[mdocod]

glad to hear that these cells are holding up to the abuse. My whole understanding of LiFePO4 when it was first coming on to the scene is that it can tolerate about the same level of abuse as a NIMH cell, can be discharged way to low a few times without too much trouble, can be over-charged on a routine basis without too much trouble, just steady wear and tear and move on with life. I think your experience more or less confirms that my thinking was correct.