My ESP32 device uses a new 18650 until it is down to 3.3 V.

My charger draws current until it is again at 4.2 V. Its indicator shows that 330 mAh has been loaded. This is surprisingly far off the 8800 mAh my cell has as nominal charge.

Is this the expected behavior? Does my device switch off too early?

  • 2
    \$\begingroup\$ mA is a measure of current. mAh is a measure of charge. Please don't confuse the two. \$\endgroup\$ Feb 26 at 10:35
  • 11
    \$\begingroup\$ Most likely someone misread 8800mWh as mAh, a common mistake from Ignorant resellers. 8800 / 3.6V = 2400 mAh \$\endgroup\$ Feb 27 at 4:18
  • 2
    \$\begingroup\$ And more, if you have the smart charger that can measure integrated charge, can you run a full discharge and charge cycle and thus determine the real capacity of your cell? \$\endgroup\$ Feb 27 at 6:12
  • 6
    \$\begingroup\$ @Ale..chenski: 3.2V or 3.3V is a perfectly fine discharge cut-off voltage. You barely lose any capacity unless temperatures are very low or currents are very high. At the same time it’s good for battery lifetime to not go lower. batteryuniversity.com/img/content/… \$\endgroup\$
    – Michael
    Feb 27 at 12:06
  • 2
    \$\begingroup\$ My experience is that going all the way down to 2.7V will rapidly accelerate battery decay. I set mine to 3.0V hard-cutoff, and 3.2V warning-cutoff (where high loads like motors would cut off). \$\endgroup\$
    – Jon Watte
    Feb 27 at 22:18

3 Answers 3


There's no such thing as 8800mAh 18650 cell. The highest capacity available is around 3600mAh. Thus your cell is fake.

Screenshot from YouTube video - C'sB Channel "Fake 18650 China batteries Slideshow"

Image source: YouTube video - C'sB Channel "Fake 18650 China batteries Slideshow"

If the label said 3000mAh or any other realistic capacity, then the cell could be legit... or fake, depending on where you got it. But with 8000mAh, there's no question about it.

If they sell counterfeit products, they may as well go all the way. Why pay for a 3000mAh cell if you're going to change the label anyway and sell it with a fake description? Something out of a dumpster will do just as well.

I'd recommend not using it, because you never know if it came from the "did not pass safety testing" dumpster.

  • 14
    \$\begingroup\$ To expand very slightly on that: the chemistry of a lithium cell quite simply can't fit that amount of energy storage into that volume. \$\endgroup\$ Feb 27 at 7:52
  • 2
    \$\begingroup\$ ~300mAh is a very common capacity for fake cells. Odds are it's a much smaller (shorter) cell inside connected up to the terminals (of course, not recommending that you open it up, as that could be quite dangerous!). \$\endgroup\$
    – Bob
    Feb 28 at 2:25
  • 2
    \$\begingroup\$ Wonder who had the brilliant idea to name a Lithium battery "UltraFire"... Unless it's a big warning of the risks associated with that battery! \$\endgroup\$
    – jcaron
    Feb 29 at 12:51

Throw this fake cell away and buy a new one from a reputable supplier. Look for capacities in the range 2,000-3,000 mAh, which is what is physically possible to achieve with current LiIon battery technology in the 18650 form factor -- anything slightly over that is likely to be fake, double that and up is 100% fake.

Be prepared to pay considerably more than you paid for the fake one. Make sure there's a return policy, so if you discover it is again fake, you can return it.

If I'm not mistaken, the fake ones generally weigh less, so if you're buying one in person, a possible way to separate the clear fakes from the possibly good ones is to bring a weigh scale and check against the expected weight from a reputable cell. This datasheet from a Samsung 2,500mAh cell indicates it's in the neighborhood of 44g.

  • 3
    \$\begingroup\$ Weight is a very unreliable indicator of whether the rating of a cell is believable or not. There's loads of examples out there of fake cells half-filled with sand to make them the "right" weight. Just buy from reputable sources, or buy no-name cells with reasonable-sounding specs and expect that you will have to test each one and toss half of them out. \$\endgroup\$
    – TooTea
    Feb 27 at 9:57
  • \$\begingroup\$ @TooTea: Does manual testing of no-name cells reduce the risk of them catching fire during use, at least after testing is complete for ones that pass? (Presumably it's important to at least be prepared for fire during testing, preferably outdoors.) \$\endgroup\$ Feb 27 at 20:49
  • 4
    \$\begingroup\$ @TooTea Weighing can't guarantee that something isn't fake, but it has the possibility to guarantee that it is, right? If it's the expected weight then you can't be sure, but if it's the wrong weight then you can be sure it's fake? \$\endgroup\$ Feb 27 at 21:27
  • 1
    \$\begingroup\$ @PeterCordes It might (statistically speaking),by eliminating the worst garbage. But probably not by much. Buying quality cells is definitely preferable. My point was that obviously garbage cells aren't an exception in the no-name space, there's a significant fraction of them. \$\endgroup\$
    – TooTea
    Feb 28 at 7:27
  • 1
    \$\begingroup\$ @PeterCordes Okay, I can see that what I meant as a word of caution against no-name cells actually comes across as a partial endorsement. That certainly wasn't my goal. I am firmly of the opinion that even name-brand lithium batteries can be a serious fire risk when not handled 100% correctly (which is what does often happen in all sorts of hobby projects) and using cells of questionable quality adds another dimension to that. \$\endgroup\$
    – TooTea
    Feb 28 at 11:47

As several have responded, this is a fake cell. The small print says 3.7V at 11.8Wh = 3200 mAh, which sounds reasonable. So @Ale..chenski estimate was right. He also confirmed my suspicion that a cutoff of 3.3V looked a bit odd to me, but that's the problem of the controller on the ESP-32 which is hard-wired.

(Added later: image) enter image description here

  • 2
    \$\begingroup\$ Dieter Menne - Hi, Re: "The small print says 3.7V at 11.8Wh = 3200 mAh, which sounds reasonable." It's not clear where these different specs are coming from i.e. the 8800 mAh in the question and the 3200 mAh you mention here. However I would not believe the 3200 mAh number any more than I would believe the (definitely fake) claim of 8800 mAh, as even 3200 mAh is very high capacity. Note here that the fake cells he tested are usually in the range 400 - 550 mAh. Even if fully discharged to, say, 2.7 V, your cell (300 mAh to 3.3V) will not be 3200 mAh. \$\endgroup\$
    – SamGibson
    Feb 27 at 15:09
  • 3
    \$\begingroup\$ Dieter Menne - Hi, Thanks for adding that image. The discrepancy between the figures for mAh & Wh is another reason not to have confidence in either of them! As I said in my comment above: Given your test results, when discharging to 3.3 V, show a capacity of 300 mAh (or 330 mAh, both figures are mentioned) then discharging all the way down to 2.7 V will not show a capacity increase of 10x up to 3200 mAh. || So yes, 3200 mAh (when fully discharged) is theoretically plausible for an 18650, but it won't be true for a cell that's already known to have only ~300 mAh when discharged to 3.3 V :( \$\endgroup\$
    – SamGibson
    Feb 29 at 13:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.