Repair 18650

[ASICS]

Three years ago, we moved to a house near my child’s school. As a result, our robotic vacuum cleaner has been sitting unused in a corner for those three years. Since we didn't remove the battery pack, it won’t charge anymore. Based on my experience, this is likely due to over-discharge from being left unused for so long, causing the batteries to enter a protection state. I removed the battery pack and, after checking, found that two of the cells had an output voltage of 0. After disassembling the pack completely and using a specialized charger with recovery capabilities, I still couldn’t get them to charge. The batteries are made by Samsung.

I'm very sorry, I don't understand how to insert pictures yet.

Under the positive terminal of an 18650 battery, there is a small vent cap. By using a small stick to press it like a button, you should feel a slight click, indicating it can recover. However, the gap between the positive terminal and the battery casing was very small, making it difficult to insert a stick.

So, I carefully drilled a 2mm hole in the center of the positive terminal with a hand drill. This step must be done with extreme caution because drilling too far could damage the battery structure and, in severe cases, ignite the battery. If necessary, it's best to handle this outside, away from flammable materials, to avoid a fire. Safety first! After drilling the hole in the positive terminal, I didn’t see the expected vent cap, just a small hole, which was definitely not drilled by me. So, I gave up on trying to save these two batteries.

After dinner, with some free time on my hands, I thought about the two batteries again. Since I was planning to throw them away anyway, I decided to disassemble them to see their internal structure. I removed the positive terminal with pliers.

At this point, I still hadn’t given up. I used a toothpick to poke the small hole in the center with some force. Then, using a multimeter, I rechecked the battery voltage. Amazingly, it showed 3.7V, indicating recovery! I did the same for the other battery, and it also recovered. It turns out that Samsung batteries are constructed differently at the positive end compared to the Japanese-made batteries I had encountered before. Although the positive terminals are gone, the batteries should still be usable. The hole in the center is large enough to use a spring contact as a replacement for the positive terminal, allowing continued use. Alternatively, you could re-solder a tab with a small hole onto the edge of the positive terminal.

To reiterate: batteries that seem dead may actually be in a protected state rather than truly damaged. If you want to restore them, you must be extremely cautious, as improper handling can easily cause a fire.

Additionally, if the batteries are indeed dead, please do not dispose of them carelessly. Treat them as hazardous waste to prevent environmental damage. Let’s all work together to leave a clean and pollution-free environment for future generations.

[Whales]

 

To be clear:

1. These are "unprotected" 18650 cells (no circuit board on the end, no flat wire down the side to the negative terminal).
2. You removed the metal from the positive end of the battery to expose a failed protection device and gain access to a working battery terminal.

I have vaguely heard that some cells have overcurrent or thermal fuses built in.  Perhaps this is what failed (possibly by unintentional means)?

[Siwastaja]

DO NOT DO THIS. Safety device (CID) disabled the cell to prevent fire. You now bypassed this device. Fire is a very likely result.

[ASICS]

You're right.

[Whales]

Is it any less safe than a LiPo pouch cell?
I was under the impression that only some 18650s have this protection anyway?

(Not that I recommend removing layers of protection!  The more the better)

[Siwastaja]

I was under the impression that only some 18650s have this protection anyway?

Every half-decent 18650 has this CID which disconnects on overpressure. There is, because it's quite cheap and simple to manufacture. China special UltraFire or something might not.

Not only it's important, here it's being bypassed once it failed, so the risk for what it exists for already realized. Now we can of course speculate that maybe the protection was being overly sensitive and the cell is still actually safe... but such speculation is getting pretty dangerous. This is like fire detector beeping on house fire, and response being assuming false alarm and silencing it, without checking if there is a fire (there likely is).

[NiHaoMike]

Then, using a multimeter, I rechecked the battery voltage. Amazingly, it showed 3.7V, indicating recovery!

Sounds like the internal fuse was blown by overcurrent due to some fault with the circuit rather than overdischarge.

[amyk]

I suspect the OP is a troll.

[Siwastaja]

Then, using a multimeter, I rechecked the battery voltage. Amazingly, it showed 3.7V, indicating recovery!

Sounds like the internal fuse was blown by overcurrent due to some fault with the circuit rather than overdischarge.

Overdischarge would also cause gassing, similar to "puffing" (in pouch form factor). The CID operates on overpressure.

Internal fuse is usually a PTC type which reacts on both overcurrent and overtemperature (more specifically, tripping current lowers with higher temperatures), but being a PTC, it self-resets when it cools down.

[NiHaoMike]

Overdischarge would also cause gassing, similar to "puffing" (in pouch form factor). The CID operates on overpressure.

But then the voltage wouldn't be 3.7V.

[MT]

Mr ASICS is definitively pulling a leg or two here!

[Siwastaja]

Overdischarge would also cause gassing, similar to "puffing" (in pouch form factor). The CID operates on overpressure.

But then the voltage wouldn't be 3.7V.

Could still be: "hidden" damage (copper dissolution) caused by overdischarge, which then materialized into overpressure on subsequent charge. Overdischarge can cause internal and partial short circuits which can also self heal. Pretty much any failure mode is possible, I would not rule anything out.

[tom66]

Cell voltage can absolutely remain present whilst the cell has experienced damage, e.g. the start of a thermal runaway event that for whatever reason did not fully propagate.

https://www.semanticscholar.org/paper/Fast-Thermal-Runaway-Detection-for-Lithium-Ion-in-Koch-Birke/f995fd9ec19bcbaa93d997d60ca6a31fddf0dd15/figure/4

There was a story of a Jaguar iPace owner who had one of the many recalled batteries (LG Chem maintaining stellar reputation here, being the source of recalls in GM and Porsche vehicles too);  he reported hearing a "popping and cracking" sound from his car and plenty of smoke.  Concerned for the risk of fire, he was able to drive his vehicle out onto the street, with a battery actively undergoing thermal runaway.  I expect it wouldn't have lasted much longer, but 3.7V does not mean a battery is OK. 

https://insideevs.com/news/601885/jaguar-i-pace-fire-while-charging-florida/