Cell reviving question

sunseeker

New member
Joined
Aug 30, 2017
Messages
16
Hi guys,

I have a question about detectingbad 18650s afterreviving them fromvery low voltage (after being salvaged from alaptop pack).
I used to kickstart an almost dead cell by connecting it to a working 4v cell for a couple of seconds.

The voltage is usually being ramped up to 2.x volt, butI can thenoftenobservethe voltage dropping by something like.01v every couple of seconds while I measure with the voltmeter.
Can Ireliably say that this is a bad cell? Should a usable cell be able tohold even this tiny bit of charge or do I need to properly test this cell?

What is your experience?

Thank you
 

neilmc

Active member
Joined
May 22, 2017
Messages
128
2.x v is still very low for an 18650. The discharge curves on them aren't linear and they have very little stored energy at 2.xv. I'd go ahead with a normal charge and test on such a cell.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,581
you should apply voltage of 3V and very low current lets say 50mA. untill it reached 3V.
When applying it quickly like you do you can kill the cell even more and yes the voltage will drop to resting after its done. its normal when doing it fast.
 

DarkRaven

Active member
Joined
Sep 2, 2017
Messages
1,235
What you are doing is the brute force method of reviving a cell. Not the most gentle way by far and not recommended if you ask me.

Your observations are correct though, this is how charging cells works. The cell voltage will go up more or less in line with the amount of current you put through it. And by connecting another 18650 in parallel this is a lot of current. The voltage will rise especially fast in the voltage range where the cell has very little capacity, that is under ~2.5V.
After the current flow stopped the voltage will drop to what is known as the resting voltage or idle voltage, which is the voltage the cell has without being affected by a charging (or discharging) current. This has nothing to do with that cell being bad, it is just how cells work.

What you should be doing instead (or what I consider to be the correct approach) is selecting the cells for reviving first. I don't bother with cells below 1.5V but everyone has their own threshold. Then you need your CC/CV power supply, like a lab power supply, or a DC/DC converter module with CC/CV function. For accuracy reasons I use a proper lab power supply.
Set it to 4.2V (or any voltage above the cell voltage actually, I'm not charging the cell all the way anyway) and limit the current to a small value, just as daromer said. If the cell is between 1.5V and 2.0V I start with 50mA. The cell voltage will rise and above 2.0V I raise the current to 100mA. I continue charging to about 3.0V, further raising the current (200/300mA) on the way if necessary (i.e. if it takes too long). At this point the cell is detected by and can be charged with a usual lithium charger. You can't use them to charge undervoltage cells right from the start because they reject them.

If it is still a good cell can be determined after the cell has been fully charged to 4.2V using your preferred device. With a maximum charging current of 1A the cell shouldn't get warm during the charge. If it does then it is a bad cell. You the discharge the cell and check the capacity. Then you charge it again and leave it for a while to check back later and find out if it has lost much of its 4.2V.
There is a way to check whether the cell is bad or not during the actual "reviving process" without having to charge and discharge the cell completely. If the cell voltage still skyrockets in the region of 3.0V although there is only a very small current flowing then it probably has very low capacity. However, this isn't completely accurate and you need a feeling for what it a fast rise in voltage and what isn't.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,581
Its worth mentioning that for instance the Opus take rather low voltage cells first and then bump charge it slowly to like 2.5V or something.
 

owitte

Active member
Joined
Oct 19, 2016
Messages
250
The TP4056 is also capable of gently charging up low voltage cells, it even takes cells with less than 1V, charges them slowly up to around 3V before pumping higher currents into them.
 

sunseeker

New member
Joined
Aug 30, 2017
Messages
16
Thank you guys,

putting a cell in parallel to ramp upa low voltage cell was more like a recent idea totest if an almost deadcell would take any charge at all.

Usually, do as I learned from the youtubers andgently start low voltage cells by putting them together in parallel and connecting them to an Imax B6 and use the NiMe program and like 1A for 12 cells to get the voltage up. From about 2.5 or 2.8v, I'll give them 200-300mA each for an initial chargeand check on them every now and then to see if they get hot.

image_khrmgg.jpg


I've tested about 700 cells so far, so I will continue to do so without the brute force method for very low voltage cells :D

Thank you guys again!
 

Terry

Member
Joined
Jul 18, 2017
Messages
58
I did a bunch of my below 1.0v cells last night, most of them were below 0.1v and the tp4056 accepts them just fine, obviously there are a few that are actually dead and can't be revived, but that's par for the coursr
 

Chris Evans

New member
Joined
Jul 4, 2017
Messages
8
I've been trying to revive as many of my poor cells as possible. In some case they seem to hold charge. What I would suggest you do is once you've revive charged them (which can normally take quite some time with say 12 in parallel at low currents) stick them in a box for a good few days and test their voltage again. You'll soon see if any are self discharging. The ones that hold their charge I then discharge test.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,581
You should test self discharge anyways. No matter if they started low or not... :)
 

sunseeker

New member
Joined
Aug 30, 2017
Messages
16
@daromer Thank you for the latest video.. I will order an appropriate resistor to set up a bump-charge holder :)
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,581
I would say that is a very simple way to bump up a cell but still you keep it a bit better than just paralelling the cells to each other causing a big current rush in worst case :)


Adding the video just in case
 

nautal

New member
Joined
Sep 5, 2017
Messages
4
owitte said:
The TP4056 is also capable of gently charging up low voltage cells, it even takes cells with less than 1V, charges them slowly up to around 3V before pumping higher currents into them.

Now I have seen this in the spec sheet :
https://dlnmh9ip6v2uc.cloudfront.net/datasheets/Prototyping/TP4056.pdf
2.9V Trickle Charge Threshold (TP4056)

Soft-Start Limits Inrush Current

but when I connect a 0.5v battery to the tp4056 I see the voltage rise rapidly with a multi-meter connected to the terminals.

1. Does anyone know what trickle charge is?
Specifically, is there a constant amount in milliamps for the tp 4056?

In my experience using a multimeter on the terminals, it takes something like:
15-20 seconds from 0.5v to 1.9v
45-60 seconds from 2.0 to ~2.8v, which seems fast.
0.02v/second from about 2.9v to 3.4v depending on the battery.
3.4-3.7v which takes around 1min.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,581
It should go rather fast from 0.5V to 1.9V. though 15 seconds seem very fast. Hook it up and use your multimeter to meassure the current on that mode.
 

TAGG

New member
Joined
Aug 17, 2017
Messages
6
For my "dead" cells (<1,0V) I put them into a holder attatched to a lab powersupply (set to 3A/4V) for about a second or two, if a cell immediately rises up to ~3V(on the display of the powersupply) they tend to be good, if a cell rises up to 4V and the current drops or it doesn't draw any current, I throw it away. Some of the cells treated this way rise to ~2,6Vand then short out, this is why it's not recommended to start with high current, but from experience i can tell you, that these cells are most of the time bad anyways (self discharge or riddiculously high IR), hence I don't care if I kill them, about 5-10% of my <1V cells have this problem.
If you want to save as many cells as possible, by all means avoid high charge currents, if you want a quick and mostly reliable way to sort out bad ones quickly my testing method has proven itself :)

EDIT: All cells above i sort in 1V increments (ie. 1-2V 2-3V and 3-4,2V), then i put 16 of the same class into a charger and let them balance for 10-20mins before I switch on the charger.
 
Top