I just pulled my first 18650's from an old laptop battery (4 1/2 years old). Voltages out of the pack: pair at 4.15v, pair at 4.09, pair at 3.95v. I've charged them all up to 4.20v now. How can I go about determining which if any of the cells are worth keeping? I do not own a hobby charger. I've read about charging to 4.20, wait 24hours, then measure voltage again and toss any that dropped significantly - how much is significant? Any other tests to run? Should I test them in a high current light (a 3A KD C8 or a direct-drive UF P60 drop-in) and check tailcap current?
I tried searching first, but didn't come up with much.
If you have a way to put a load onto the battery and measure the voltage while it is delivering the current, you can get an indication of the cell condition. If I put a freshly charged cell on a 1 to 1.5 Amp load (I use a 3.3 Ohm 5W resistor) and it immediately (5-10s) drops down to less than 4.0V, it’s got problems. An good 18650 cell should stay above 4.0V for a while, maybe a minute or two anyway.
Here’s the no load voltage, 4.16V
And here’s the same cell with a ~1.2A load. This drop is borderline, and the cell has a capacity of about 1100mAh from my hobby charger test. The original rated capacity was 2200mAh.
I test this way and with a hobby charger (Accucel 6). I have two sets of pulls, both appear to be Sanyo. One set have good capacities and reasonable internal resistance. The others are not so good, so I will probably dump them when my new battery order from FT comes in. A third set produced all duds.
Thanks relic38. I'm just not sure what to use for a load, perhaps a P60 module pulled out of the host, but I'd have to set it to medium mode for a +/- 1.0 to 1.5A output. Wait, the P60 might be direct-drive in high so medium isn't current regulated and might not be 1.0 to 1.5A.
Using medium on the P60 pill could cause a problem with the meter voltage readings (this is meter dependent, some are better than others).
Go with it on high. Try it on a known good cell and note the drop. Then try it on the bad one and see how much it drops there. I think testing at 3+ amps could be a problem though.
if you have an AK-47A driver board (or generic equivalent), those are perfect for this test. use any LED star on a pill with this driver and it should pull about 1.05A.
Ok, I'll give it a try (with 3A+ P60 on high, don't have a 3x7135 board handy) if I can get time to do it. Sounds like it would be helpful in a case like this to make up some wire leads with magnets attached at the ends so I don't need 4 hands to hold things!
Yup, I have those. for the shot I took, I just clamped the wire to the - end with the magnet. That allowed me to hold the camera. My meter leads are hook style clips so I don’t need to hold those
Easy way is to just put it in a light and run it on high for a couple minutes and then test the voltage drop.
After that, I just run the battery in a light to see what kind of run time it will get versus a known good battery. I do this using the light as my room light while doing something else rather than use the energy in testing.
If run time is very or fairly short, it’s less useful and either I find a light with less draw where it’s OK, or recycle it so as not be further bothered by it.
At this point, I have so many pack pulls (200?) that I might have to start giving them away.
Testing load is fine. But for me the acid test is an actual use test, but more importantly, how well the cell holds a charge. Set it aside for a month and see what the voltage is. I have samsungs and panasonics that will hold a 4.20 volt charge for three months. I also have some that won’t hold it over night, although usually not with the panasonics. They are truly amazing batteries.
Load does not mean that they will hold a full charge worth beans. It’s an indicator that they might, but I have probably 20 Samsung grays that test great in a run test, but just can’t hold a charge. Samsung got it right with their pink and purple 2600 and 2800 18650s, but the earlier gray and turquoise ones? Not good. At least in my experience. Maybe 1 out of 5 will hold a charge. Who wants to pick up a supposedly fully charged battery, only to find out it’s 3.85 volts after three weeks on the shelf? Off to the recycling center it goes!!!
I also have several lowish power torches that I use for testing. They don’t get hot no matter how long they’re left on while tailstanding. Take a known NEW panasonic at 4.2 volts. Run it for 30 minutes. Then take the voltage. Should be about like 3.9 volts. 4.0. Something like that. Then test one of your pulls. See what the voltage is after 30 minutes. Give you a good idea how they hold up under load. The results might surprise you. I have some 2000 mah LGs that should be average performers, but they’re right up there with some panasonics. I also have some lime green OLD panasonics, some of their first li-ions, supposedly 2000 mah. They’re also killer cells.
That’s my .02. Hope it helps.
Oh, and you may have to cycle a pack pull maybe 10-15 times to get it stable so that it will hold a solid 4.2 volts for a while, at least in my experience. I’ve had losers turn to winners after using then a bunch of times, so give ’em 10 cycles before tossing them.
Hope these are good. First cells I’ve had drop much in the first day after charging, but they’ve held up with no drop in the next 3 days. Got these as my latest pulls from my own laptop. We got it for my son in about 2004 so they’ve been around a while. And they sat in storage most of the last year.
PS - LOL, your $0.02 is under-rated. I learn (maybe re-learn) every time you post on salvaged cells. That is a lot of pulls! Guess you’ll have to buy some more lights to store them.
I’ve only opened ~6 free packs and see no need to buy any at this point. My son is sending over 2 more Dell packs; he also got the laptop but the screen’s backlight doesn’t work so I’m not sure if I’ll harvest both packs, yet.
I use protected cells in multi cell lights and I will maybe throw a protected in my EDC 2011 sometimes because it has a tendency to turn on in my pocket, although I don’t know what a protected would do to protect me in that circumstance.
But other than that, I use all pack pulls pretty much all the time. (WARNING: This is NOT for everyone.) I’m sure I’ll have an incident at some time with nonprotected cells, but haven’t had one yet. But I continue to have problems continually with protected cells. Just bought four from a well-regarded web company, one of the better ones. One was DOA. Still going back and forth with the company that this is somehow my fault. A 25% failure rate ain’t good. And I’m supposed to trust these circuits? No thanks. Okay, I have now stepped off my soapbox.
Plus, with pulls, I LOVE getting stuff for free. I rarely pay for them. If I do it’s only a couple bucks. I just got like 8-10 computer packs from a home depot battery recycle bin about a month ago. My best haul yet.
What i really want to figure out how to do is how to test the pack without opening it. So if goodwill wants a couple bucks for it, I can use a DMM to test — say — tabs 1 and 7 and that will give me the overall voltage or something. No luck yet. Next pack, I’m really going to sit down until the sun comes up and figure out how to test the voltage externally. I also should look on the web. Your computer monitors the voltage, so there has to be a way to do it.
At higher current draw, the protection is stealing precious power from you. those 1cell XM-L lights running at 2.8A? there’s probably 200mV drop on that PCB and the tabs. That’s almost 600mW wasted in the battery (that’s about 8% of a 2600mAh cell, BTW).
No thanks, I’d rather have unprotected and check/charge my cells after each use.
Agreed, not for everyone, especially casual users who treat them like Alkaline or NiMH.
Cheap, or cheaply-made, protected cells frankly scare me more than unprotected cells. That sharp edged strap running over the edge and down the whole length of the cell... scotch tape? Really? What happens if you drop it and it lands just the wrong way... POOF.
Agreed, it reduces the risk of one or two failure modes, and adds in one or two different ones. I will say that the tape used for these applications (known as Kapton) is quite durable, it still could be compromised with a sharp blow.
I think once I get my real cells, I might do some “failure mode” testing on my aged pulls to see if the manufacturer-included protections (PTC and supposed puncture protection) really can prevent catastrophic failure. I’ll start a thread asking people for acid tests they want to see done.
