Cells 221 to 230 has at least 84% of its original capacity. 227 to 230 are real fresh cells.
hmm… try get some fujitsu or acer… they use panasonic 6 CGR 2900 mAh batteries for less than $20 and if you lucky you can get one from acer for $13/shipped
While I'm not a battery expert, and hoping that someone more knowledgeable than I would correct me if I'm wrong, my readings indicate that an 18650 cell should be retired when its charged capacity drops to 80% of its original capacity.
Using that criteria, and assuming that the unidentified cells are 2200mah (the most common size for laptop cells).
Cells 201 to 219 should be retired, cells 221 to 230 are good, 231 & 232 retire, 232 good, 234 retire, 235 good (just), 236 retire, 237 to 251 good, 252 retire, 253 and 254 good, 255 retire and 256 is good.
It would be really great if someone could put together a list of preferred laptop battery packs to get. There are so many different models and versions out there, it’s hard to know what’s best.
Also, it seems like it’s a real crap shoot on how much usage a battery has been subjected to. While a higher mAh rated cell is of course preferable, wouldn’t it be better to get a 2000mAh cell that’s almost new versus a 2900mAh cell that’s just barely over 90% capacity?
“recycle those perfectly matched 1700mAh Sanyos” - LOL, what a moron
What a beautiful list I must say - all those colors… like a piece of art. 
I wouldn’t recycle any - even those around 1Ah are good in 3W torches and in those that get rarely used. And others, bit better, can be gifted to non-flashaholics.
Especially those CGRs which have very low resistance even when old.
All those cells are superior to any scambay Crapfire. Simply because they are the same cells without spending few years smashing around in garbage containers.
An IT department of a local organization is constantly giving me old laptop packs, I feel no need to buy 18650 cells regardless of performance.
Hahahahha, ok ok, maybe I did overdo the colors. So should keep them all, use better ones for XML lights and those around 1.3-1.5Ah for XPG?
Some of us where buying brand new never used older laptop packs that contained Panasonics 2900mah cells. They where 3 1/2 years old and where at a very low voltage when received. When they where charged and tested they seemed to act as though they had just rolled of the assembly line. They tested like new, hold voltage over time like new. So you maybe able to get the best deal if you keep watching for some older pack that’s brand new and never used. Any OEM name brand pack should contain quality cell’s like Sanyo’s or Panasonics.
The colors perfectly underline different pack making it clear and simple, no irony at all.
That’s what I would do, but to be exact it’s more the internal resistance that should determine the use of the cell. I’d love to check all those cells by special multimeter to find out which brand keeps the lowest resistance. Having so many pulls you could consider getting this device.
By measuring resistance, you can sort out lemons without having to test capacity and also save some lower voltage cells that are not damaged.
You can get rough estimate of IR by how much current the direct-driven LED draws. Having the same resting voltage, you should be getting different amperage according to cell resistance.
But to keep it simple, the capacity deteriorates hand in hand with capacity most of the time, so being all these cells branded, the differences should be marginal.
I found this on ebay but 57$ seems rather steep, though HKJ has a nice thread here on how to measure resistance. I’ll definitely give it a try.
OK, looking forward to see mOhm column in that beautiful chart 8)
last word for checking resistance: In order to get most accurate results, you have to press the probes firmly on the cells to reduce additional contact resistance. Measuring the same cell to see if you get (the same) valid results should help you doing it right. The lowest value it shows is the right one.
Yes indeed. I would only toss the cells that will not hold a voltage or will not take a charge. Most cells I've harvested from laptop packs are quite usable, even when years old. I think this is because in a laptop pack they are charged at relatively low current. If you look at the data sheets you will find that most manufacturers suggest charging at less than .25C, and recommend such low rates for optimum charge capacity. Charging rapidly with 1C or higher rates, while quite doable, results in a less complete charge and over time, shortens the useful life of the cell.
Charging at .25C can take a little longer, but charging several cells in series compensates for that quite nicely.
neat, seems to match my experience with pulls - most of them that are 3V or more end up with 1800-2100mAh capacity. TBH I now tend to test them as parallel pairs to save time (doing single cell tests on 6-8 packs is pretty tedious) as usually one bank has bricked the pack and has <3V. I got some beautiful 3Ah lipos from some Macbook pro packs though
Are you doing a discharge capacity test on a pair of cells configured as P2S1?
yep. I used to separate parallel pairs, but they always came out within a hair of each other, so now I discharge (and self discharge) test them as 2P1S pairs. It’s much quicker to weed out low capacity cells that way. I also sometimes use them as parallel pairs anyway, so this saves me having to solder the cells or use a holder.
I guess it’s best to stick with 6 cells or less and try for 5800mAh or higher for the better deals. I was just hoping that there are some particular brand/model packs that have been identified as worthwhile to buy…
As it stands right now, I have 3 good working Samsung 18650’s that I pulled from a prior laptop battery I’d been using (went bad on me after 1 year) and I’ll probably be switching out my current laptop battery in ~6 months for harvesting another 3-4 cells. Nice to recoup a little of the battery cost. 
Interesting. It seems logical that a 2P1S pair would either be both good or both bad, but until today I have always separated them and tested individually. This evening I took a pack apart and two pairs had good voltage while the third pair had about 2.3v. I will test it as a pair and see what happens.
it’s not impossible that you’ll get a pair with one good cell and one duff one that will float after it’s charged/ discharged and there’s always a risk that you’ll discard that good cell as the parallel pair will have a much reduced capacity, but I haven’t seen that yet in the 60 odd packs I’ve torn down, so I don’t think it’s much worth worrying about. Plus, after a while, you’ll most likely end up with more cells than you need so missing out on one or two isn’t a big deal :).
In tearing down packs what I see is either one or more pairs are duff or that every cell in the pack has low capacity. The former more than the latter, probably around 2:1 or 3:1.
Thats my experience as well. I wish I included the cells I threw away in the table above, just for an illustration.