Strange charging curve on harvested cell???

Wulffi

New member
Joined
May 8, 2019
Messages
9
Hi @ all,

I started to harvest some cells out of old laptop batteries to be able to build up a small powerwall.
I will post some more information about that on a separate thread.

But right after the start, I got a cell with a very strange charging curve so that I stopped using this one and wanted to ask you.
As you can see in the attached file, the voltage has dropped several times during charging.

This cell is marked with "LGDB118650".

image_yayzmx.jpg


I am using a SKYRC MC3000 with the following setting:
Charging up to 4.20V @ 1.0A
Resting for 20mins
Discharging down to 3.00V @ 1.0A
Resting for 20mins
Charging up to 4.20V @1.0A

The battery did not get warm but I am a bit afraid that a cell could catch fire.

Daniel / Wulffi
 
jumping from 3.7 to 4.1 leads me to believe there is high resistance.
could be contacts,cid,or a bad connection in cell.
if no others exhibit this scrap it.
 
Thank you for your reply.

My idea was that jumping from 3.7V to 4.1V is just because of the starting charge phase with 1.0A and a pre-charged cell.
I was more worried about the voltage drops afterwards.

The contacts are just fine and cell resistance was approx. 300mOhms.
 
What is the discharge capacity of that cell ? (measured not from the original spec)
 
I did not do any discharge until now as I stopped the charging for security reasons.
I am a bit afraid that a cell might catch fire, so that I wanted to be on the safe side and ask someone here before.
 
Wulffi said:
Discharging down to 3.00V @ 1.0A

You mentioned this ..... perhaps it was just a manual discharge ?

It wont catch fire at 1A (probably)
 
@Sean
Yes, that is correct but I did not do the full test procedure and stopped while the first charing.
Therefore I do not have any capacity.

@kc8adu
Thank you - I will scrap this battery.
 
Toss it. You will sleep better knowing that a questionable cell was not allowed in your pack, where as if you use it, you will always wonder when/if that's the one that will melt down. I have a specific category of failure called "Trust", if I don't trust it I won't use it no matter how it tests.
 
Back
Top