I use my NCR18650B for my K40M!
BTW the GA has a higher voltage at 4.35v so capacity will be lesser than 3500mAh unless u have a hobby charger that can charge the cell to 4.35v
You don't need a hobby charger, there are several 'regular' chargers that have 4.35V settings. e.g. Xtar SP1 & WP2H
I also have some 18650B 3400 mah. Great batteries
Indeed.
This Panasonic datasheet also quoted 4.20V, so where is this 4.35V comes from @Wrathbringer?
I thought only some LG/Samsung cells are 4.35V?
:quest: :quest:
I don't see them rated at 4.35v anywhere. FastTech's said they tested at 3518 mAh from 4.2v down to 2.7v and pretty sure I got similar results when I tested capacity on them.
The PanaB's were state-of-the-art at the time, actually for a pretty long time they were the reigning king of capacity/price. Now, the GA and MJ1 trounces the PanaB's - I consider them de-crowned, obsolete, yesterday's best, etc... The GA and MJ1 is higher capacity and higher amps rated, so, they also have lower internal resistance, plus they are available at great prices (prices wars between GB and BG for example), and available with protection circuitry (MtnE and FT under the EVVA label).
The technology has passed the Panasonic brand - of course they own SANYO batteries since 2008, so SANYO or Panasonic - doesn't matter.
That Nkon link has a review where the GA and MJ1 are compared for capacity and IR.Both are 3.5Ah but the GA measured about half the internal resistance of the LG.
“The reviewer”: Seems to be better than LG MJ1 Review by Mr Nerijus Bosas Rating Tested two LG MJ1’s against two of these Sanyo’s, which both should be 3500mah capacity according to specs. Used Opus BT-C3100 v2.2 charger. Multiple times i charged them to 4.2V and then discharged to 2.8v. For the 1A discharge both cells showed almost exactly 3.5AH with no noticeable difference between two. What i did notice is that my opus charger shows MJ1 batteries to have 20miliohm higher internal resistance than Sanyo’s. Using 1khz 4 wire professional battery IR tester, Sanyo’s have internal resistance of 28miliohm and MJ1s around 50miliohms. For whatever reason MJ1 internal resistance fluctuates much more than Sanyo’s which kinda implies their might have different chemistry.I don’t like that. Based on my test i highly recommend Sanyo’s instead of MJ1’s and get yourself proper Li-ion charger. My recommendations are either XTAR SP1, if you want to charge just 1 cell and OPUS BT-C3100 v2.2 if wanting to charge 4 or less cells at same time. The later one also shows battery internal resistance and can calculate battery’s capacity while discharging with up to 1A and charging up to 2A. Just notice it has internal cooling fan so can be annoying in very quiet enviroment…Hope my review helps you guys…. (Posted on 6/23/15) ……seems to know what he is talking about 
HKJ’s reviews will be the acid test though.
Y’alls batteries are sooooo…like…yesterday’s tech.
I am in possession of UltraFire 18650s that are 6000mAh. It says so right on the label!
I think it was 30b my bad sorry
I hear you krono 
. I just conducted some unofficial comparison tests between the MJ1 and GA, all with a FET based driver in a modded ZY-T11 clone with a T4/T5 5B1 XM-L2. With two MJ1's and two GA's, all charged to the same voltage level of 4.21v +/- 0.01v.
The MJ1's show consistently higher tail amps measured: MJ1 avg: 5.25A, GA avg: 4.84A, but outputs on two 30 second runs on each cell shows very similar discharge levels and lumens, well under any margin of error, typical lumens over 30 secs is:
1115 @start, 1081 @30 secs
This is interesting, because they show different characteristics with the tail amps, but real output differs very little. Of course this is not as scientific/thorough as HKJ's, and I'm only covering the first 2 minutes of discharge roughly at 5 amps.
With the same LED and torch,you would need a flat lumens vs amps graph,or be either side or the peak lumens to get the same light output at different currents.Djozz’s graph shows the XP-L peak around 7.5Amps,not 5 :~
And I thought QTC’s were black magic!
Wut? I don't understand. What is QTC and why XP-L?
Dunno, but for my tail amps measurements, I remove the tail, then for lumens testings, I re-assemble the tailcap on the light. It's a difference, and maybe the cells differ in handle the resistance change: DMM vs. tailcap (with spring bypass of course). Again, my tests are not on a bench with a bench PS or anything.
It's all a big guess until HKJ tests them, I've always found his internal resistance results seem to match up with my own "in light" test results, so more waiting til he gets them in.
Edit - I see he ordered them here: https://budgetlightforum.com/t/-/34262
For some reason,I thought it was a Manker A6 on copper,not,as you have clearly written,a ZY-T11 with XL-M2 T4/5.I also imagined the torch in the light box/tube with the tailcap off and the DMM connected.My mistake.
QTC-quantum tunnelling composite,spiky zinc nano particles in a silicone matrix,the more it is squeezed,the lower the resistance,from open circuit to a dozen or so milliohms.The spikes don’t touch,but as they get closer,they allow a “potential” for lower resistance.I’ve measured kohms across a QTC pill driver while seeing several dozen lumens out the front of a 16340 sized torch.Black magic!
Testing one pair of cells against one other pair is a bit small for a sample size. Gross differences will probably remain consistent but small percentage differences could be lost in averaging a larger sample size. In other words knock yourselves out but don’t take it for gospel until it’s been repeated many more times with many more pairs. I appreciate HKJ’s tests and refer to them often but he doesn’t pick cells at random from the assembly line, they are sent to him and by now they might be aware of the tests he publishes online and could be more likely to send him a less than random sample, who knows? I’m sure each plant does it’s own in house QC sampling but how that compares with what they publish is anyone’s guess. Just some less than random thoughts and opinions with absolutely zero data to back it up.
Quite possibly grade A cells.
Ahh - might have been true in the past, but in that HKJ thread, he asked us for what cells we would like to see tested, and he went and ordered them himself, so much better this time. Also in the past, dealers not manufacturers sent him cells, so again, it's more like picking them off the production line. I know it's not for all cells, all cases, but definitely some pairs of cells were not, could not have been hand picked in his testing.
He is unbiased in the testing, but still needs cells to test, and can't afford always buying them himself. So, you can see the pattern of cells he tested, many with the same label, some just re-wraps, but they were the ones that sent him the cells, not us members here at BLF or CPF, for example.
Boy, I sure do wish we could get a collection going for testing - to fund HKJ and dzozz with the latest stuff. I think it would be money well spent.
+1
Several valuable members who do a lot of testing that benefits the community, should set up a https://www.patreon.com/ account.
I would gladly contribute to members who invest there time & effort for the BLF cause 
and expand our collected knowledge we all stand on.
2 cells i have found in posted testings here at BLF, often shows an higher amp than the lumens output will show compared to other similar cells is the Sony VTC5 & Samsung 20A.
These kind of differences have always (since i noticed them in your guys testing) puzzled me, i suspect it is because the volt drop is less in the cells who shows lower amp but still similar lumens output. Or maybe the different chemistry’s are pulsing in output at different rates.
Because when i run lights DD without a driver, i often can get a pulsing effect if i set up a twisty, so that it get less connection the last fraction of a mm to “make” modes by very delicately controlling the twistys rotation before full connection, often it gets in to a sort of feedback loop of pulses particularly at the lower output levels close to moon & firefly.
I think the issue was/is when will it makes sense to change the cell used in “standard” testing.