Its called Quality Fade 
I heard back from one mfr, and yeah, they were being cheated so are moving to a different supplier.
First was an Anekim 18650, 3000mAH, that clocked in (see my UC20 review) between 2300 and 2400 instead.
Just this weekend, I got a Wurkkos 21700, 5000mAH, that clocked in at around 4600 and change instead. Looked twice to make sure it wasn’t supposed to be a 4000mAH instead… nope, 5000mAH.
And a Sofirn 16340, 800mAH, that only clocked in at 460-470 (ran it twice, exact match for both runs!). Yeah, little more than half the capacity.
Okayokayokay, could my Ope be going mental on me? Stuck in one of my reliable 30Qs (3000mAH, of course), and it clocked in at 3040 or so. Definitely not the Ope reading low.
Before, every single Sofirn/Wurkkos cell I got, clocked in at pretty much exactly the rated spex, never even had to think about it. Wondering if they all share the same supplier who’s playing games with them all. If so, they gotta dump that supplier before they get burned as far as reputation.
(Wondering if Fun’s rusty cells could be related to this, and be from the same supplier.)
I doubt that. Sofirn/Wurkkos get most of their 21700 batteries from Lishen, 18650 batteries from DLG. Some of their latest 26650s come from PLB. It is very unlikely that all these battery manufacturers have started some conspiracy against flashlight manufacturers. What I see over here happening is:
- battery shortage is getting more serious day by day
- demand on the other side is rising (EVs, notebooks, power tools, ...)
- more and more fake or grade B cells showing up, even sold from trustworthy retailers (by accident I suppose)
My assumption is that not only Samsung, LG Chemicals, Sanyo/Panasonic, Sony/Murata but also Lishen or Molicel have to deal with huge demands and limited stock availability. Top grade cells will most likely be sold to EV manufacturers who buy high volume lots, being considered as premium customers. The flashlight/vaper market is insignificantly small in comparison.
Lots of sad truths in that post above from lux-P. I don’t have a discharge function on my Nitecore charger but it does measure how many mAh goes into each cell. The other day I ran 4 Sofirn 18659 3000mah cells down to 2.9v (I was using an MF01S as a lantern) & when fully charged at 4.25v they recorded high 2900’s some over 3000. I don’t know that my $35 Nitecore charger is a benchmark of accuracy though.
I had an issue with one of the 26650 cells included with the DL70 I reviewed. Aside from being advertised as 5000 mAh cells and testing as 5500 mAh cells, one cell probably had high internal resistance and sagged faster than the other (the DL70 is 2S). After the runtime tests, one cell was still over 3.2 volts but the other was discharged really deep, like under 2.6 volts (it gradually went back up to 2.8 volts though after resting a few minutes).
Is this a sign of mismatching cells from different productuon runs and quality? Maybe, but with the battery situation in China like it is, hard to speculate we will see more of this.
I will test my sofirn cells, when my 2nd order arrives. I have bt-c3100s. That will span 26650s (dl70) 21700 (ts21) and a couple 18650s on the way.
I wonder how they bundle cells, probably as needed, first in first out?