18650 high self discharge cell becomes fine again?

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
Does anyone know why a "bad" 18650 cell, suffering from very high self-discharge, would suddenly hold onto a charge like it was perfectly healthy?

I have a button-top Samsung 30Q cell (the button top was added after-market) that is only a few months old. I tested it as a good cell when I got it, and it would hold a charge fine. About a month ago, it started suffering from very high self-discharge. It would drop about 0.1v per day, sitting alone on a shelf. I obviously got concerned about it, and figured it must have a small internal short or something like that. When I charged it, it still had its full capacity, the only problem was that it wouldn't hold onto a charge very well.

I decided before getting rid of it, I'd maybe use it for some experiments or something that I wouldn't use a "good" cell for. I started charging it every day, to top it up (while in a metal can for safety). It would always drop to around 4.07v by the next morning. If I left it, it would continue to drop at a steady rate.

Suddenly, I notice that it's no longer dropping voltage, and is holding steady at 4.17v. Huh????

Does anyone know why a cell would go from "good" to "bad", and then back to "good" again?
 

JoeRodge

Newly Enlightened
Joined
Oct 8, 2017
Messages
133
Maybe it was some form of a memory effect.

Did anything different happen this time? Did u let it discharge further before charging it?
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
It was from a matched set of other cells, and all the others (used in the same light) are fine.

I didn't do anything different the last time I charged it. I have been topping it up every 1-2 days, just to measure the voltage drop. Then, suddenly, it stopped dropping by its usual 0.1v per day. It has dropped 0.005v in the past day, which is a little fast, but way better than it was doing.

It was a real loss of energy. When I first noticed it at 3.5v, it took 2000mAh to charge it back to the full 4.2v. So, the drop in voltage is resulting in a real loss of energy, it's not a weird resistance effect or something.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello WalkIntoTheLight,

Let me speculate...

If a cell develops the start of an internal short, you would notice a higher rate of self discharge. Through "exercise" you may be able to dissipate the tendency for the internal short and see things return to somewhat normal.

If it were my cell I would recycle it.

If you want to continue to play with it, use caution during charging, discharging, and storage.

Tom
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
Hello WalkIntoTheLight,

Let me speculate...

If a cell develops the start of an internal short, you would notice a higher rate of self discharge. Through "exercise" you may be able to dissipate the tendency for the internal short and see things return to somewhat normal.

How would that work internally? I don't understand how a short could recover. If anything, I expected it to get worse.

If it were my cell I would recycle it.

If you want to continue to play with it, use caution during charging, discharging, and storage.

Tom

Yes, I think you're right about recyling it. I still want to see what happens to it, before I get rid of it. I keep it in a tin can at all times, as well as during charging.

So far, it's acting like a normal cell again. It's kind of scary to think that a cell might have an internal short, but somehow recover so you would never be able to tell.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello WalkIntoTheLight,

In the days of NiCad dendrites would form and then through "exercise" be burned off. This delayed the death of the cell but did nothing to improve its overall health.

Li-Ion chemistry is different but it is possible that the chemicals can shift around during "exercise" to change things a little. Usually internal shorts are immediate death but it is possible that something out of the "usual" can occur.

If you search on the degradation of Li-Ion chemistry you can catch a glimpse of some of the chemical and physical reactions involved. Internal shorts usually involved a compromise of the separator but may also involve a breakdown in the chemicals of the electrolyte.

Tom
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
In the days of NiCad dendrites would form and then through "exercise" be burned off. This delayed the death of the cell but did nothing to improve its overall health.

Li-Ion chemistry is different but it is possible that the chemicals can shift around during "exercise" to change things a little. Usually internal shorts are immediate death but it is possible that something out of the "usual" can occur.

If you search on the degradation of Li-Ion chemistry you can catch a glimpse of some of the chemical and physical reactions involved. Internal shorts usually involved a compromise of the separator but may also involve a breakdown in the chemicals of the electrolyte.

I first suspected it was a damaged separator, but that wouldn't explain why it got "better". Maybe it is something chemical, as you say. But right now, the cell still appears to have full capacity. I'll do another check on it sometime in the next few days.
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
I've ran this cell through a couple of full charge/discharge cycles. It's about 2750mAh to 2.9v, which is near what my other 30Q cells do. So, there doesn't appear to be any loss of capacity, or ability to do high-discharge (in FET lights), and it's holding a charge now.

Maybe there was an internal dendrite grown through the separator, but it has been broken or re-absorbed. Can that happen? If they can be fully reabsorbed into solution, maybe it's okay?

I'm a little concerned that a cell can apparently go bad, but then fully recover. If I didn't measure this cell during the couple of weeks it was self-discharging, I'd never have known.
 

VapeMagoo

Newly Enlightened
Joined
Jun 10, 2018
Messages
1
Does anyone know why a "bad" 18650 cell, suffering from very high self-discharge, would suddenly hold onto a charge like it was perfectly healthy?

I have a button-top Samsung 30Q cell (the button top was added after-market) that is only a few months old. I tested it as a good cell when I got it, and it would hold a charge fine. About a month ago, it started suffering from very high self-discharge. It would drop about 0.1v per day, sitting alone on a shelf. I obviously got concerned about it, and figured it must have a small internal short or something like that. When I charged it, it still had its full capacity, the only problem was that it wouldn't hold onto a charge very well.

I decided before getting rid of it, I'd maybe use it for some experiments or something that I wouldn't use a "good" cell for. I started charging it every day, to top it up (while in a metal can for safety). It would always drop to around 4.07v by the next morning. If I left it, it would continue to drop at a steady rate.

Suddenly, I notice that it's no longer dropping voltage, and is holding steady at 4.17v. Huh????

Does anyone know why a cell would go from "good" to "bad", and then back to "good" again?


I seem to be having a similar problem, but without any sign of recovery yet. It was one in a pair of PVA Type B rewraps (noted at one point by Mooch as a 30Q and they do resemble and have parameters that match other 30Q's I have from NKON).

Charged yesterday to 4.20v and in about 12 hours it has dropped to 4.09v. I did charge it before to 4.20v and let it set about a week and it dropped to 3.41v. I have run it through a couple discharge/recharge cycles in the charger but no improvement. I also noticed that it drops to about 4.15v within 2/3 hours after charge. I tried in 2 different chargers to see if it will top off from 4.15v. Both chargers sense it as FULL quickly but checking the voltage on the chargers it is still at 4.15v. Just to see what happened I tried to charge it along with its mate in a mod. The mate had no problem, the mad cell seemed to be stuck in a perpetual charge at 93%.

This cell has never been abused and has been used less than 10 times, less than 15 times counting charger testing cycles.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello VapeMagoo,

Welcome to CPF.

Obviously something has caused internal resistance to increase inside the cell. This is not good and may not be safe.

Time to recycle the cell and replace it.

Tom
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
Eventually, mine went back to being "bad". Not as bad as before, though. After charging fo 4.2v, it seems to hold a voltage around 4.02v if I leave it for a week, with most of the voltage loss within the first day or two. Before, it was dropping by 0.1v every day, all the way down to the the low 3v range (before I'd charge it up again).

Internal resistance doesn't appear to be an issue. It's about the same as all my other 30Q's. (I don't trust my charger to give me accurate measurements, but if I average enough of them I get a sense whether something is high or low.)

It was a button cell, so I unwrapped the outer wrap and pulled off the button to make it a flat top, just to see if it was something to do with that. No change, however, other than seeing the IR go down a bit (no button to add resistance).

Weird.
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
I had another button-top 30Q cell go bad recently. Same thing as my first one last year. High self-discharge starts to occur, but other things like capacity and internal resistance appear to be fine. A third button-top 30Q is starting to go that way too, though it's only dropping a few millivolts each day. I expect that will speed up.

This is annoying. I use these cells in a 4x18650 light in a series configuration, so it's really important that cells maintain consistent voltage.

These cells were all purchased from Banggood. I thought they were genuine Samsung 30Q cells (with an added after-market button), but now I'm not so sure considering their failures. Flat-tops I've bought from Banggood all appear to be fine. Maybe whoever is supplying them with the button-top versions is using fakes?

When I remove the outer wrapping, they are marked like genuine 30Q cells, but that doesn't mean much.

Anyway, thought I'd warn people to check their button-top 18650 cells, especially if they bought them from Banggood within the last couple of years.

Out of 9 button-top 30Q cells I bought there, 3 have gone bad (or are going bad). That's not a very good failure rate, IMO.
 

ChrisGarrett

Flashlight Enthusiast
Joined
Feb 2, 2012
Messages
5,725
Location
Miami, Florida
I had another button-top 30Q cell go bad recently. Same thing as my first one last year. High self-discharge starts to occur, but other things like capacity and internal resistance appear to be fine. A third button-top 30Q is starting to go that way too, though it's only dropping a few millivolts each day. I expect that will speed up.

This is annoying. I use these cells in a 4x18650 light in a series configuration, so it's really important that cells maintain consistent voltage.

These cells were all purchased from Banggood. I thought they were genuine Samsung 30Q cells (with an added after-market button), but now I'm not so sure considering their failures. Flat-tops I've bought from Banggood all appear to be fine. Maybe whoever is supplying them with the button-top versions is using fakes?

When I remove the outer wrapping, they are marked like genuine 30Q cells, but that doesn't mean much.

Anyway, thought I'd warn people to check their button-top 18650 cells, especially if they bought them from Banggood within the last couple of years.

Out of 9 button-top 30Q cells I bought there, 3 have gone bad (or are going bad). That's not a very good failure rate, IMO.

'I'm going to go with 'what are fakes?' for $500, Alex.'

Chris
 
Top