Review and test of batteries from nkon.nl: Samsung INR18650-30Q

It's been a while since I reviewed branded batteries. Thanks to regular reader Vladimir, I now have a subject for this review—or, more precisely, for an entire series of reviews testing branded batteries. And now I also have proper equipment to test them.


Before starting the review, I'd like to thank Vladimir for providing the batteries. This isn't the first time, either: about a year ago I published a series of reviews of Mean Well power supplies, and I owe him a huge thank-you for those as well.

Today's review and the ones that follow will cover branded batteries. I used to buy them from Gearbest, and I've reviewed some of those batteries, but in practice there aren't many places where you can buy genuine batteries.
These include nkon.nl and FastTech.com. There are probably others, but the chance of getting a counterfeit is minimal at these three stores. That said, Gearbest's policies have changed somewhat recently, so you need to pay closer attention to the listings.

I received two small boxes containing the batteries. The packaging isn't retail packaging; I just liked how compactly everything was packed :)


I received nine batteries in total, of five types. All except the Samsung 21700 came in pairs, since one isn't enough to draw any conclusions, while two already give you some statistics.


I'll start with the Samsung INR18650-30Q. It has a rated capacity of 3000mAh and can handle currents up to 15A, although I'll add a small caveat about the latter later.

The color, smell, taste—everything is genuine. The production date on these batteries is printed on the casing, not on the heat-shrink wrap.
It's hard to see, so I'll repeat the main marking here:
L0HB
6BC31

The problem is that these use a different code from the batteries whose code is printed on the heat-shrink wrap, so using the old decoding method gives nonsense.
For newer Samsung batteries, where the code is under the heat-shrink wrap, the following decoding applies:
L - production line number
0 - model number
H - year of manufacture: Q-2008, S-2009, Z-2010, B-2011, C-2012, D-2013, E-2014, F-2015, G-2016, H-2017, J-2018, etc.
B - month of manufacture: 1-January, 2-February, 3-March, --- A-October, B-November, C-December

Where the marking is printed on the heat-shrink wrap, the same letter/number mappings apply, but they appear in a slightly different order—not
Factory line number
Model number
Year
Month

But rather
Factory
Year
Month
Week

The marking is barely visible in the photo, but it's definitely there :)


Dimensions and weight. At this point I immediately labeled all the batteries 1 and 2; they'll keep those numbers in the tests to make comparisons easier.


The battery's specifications, according to the datasheet I found on the same Nkon website.


According to the datasheet, the resistance at 1kHz should be 13.1mΩ ±2; in reality, we get 12.8–12.9, which is excellent.


For the first run, and to get an initial capacity measurement, I use a simple charger that can also measure capacity accurately during discharge.
Charged at 1 A, discharged at 500mA. The figures on the left are from the start of the test; the batteries arrived at about 25–30% charge.
The results were 3186 and 3160, for test subjects 1 and 2, respectively.


Next come the more accurate tests.
A small caveat: by chance, I initially used a different datasheet, which gives slightly different charging parameters.
The datasheet from nkon specifies a charging current of 0.5C (1.5A) and a cutoff at 100mA.
The one I found earlier specifies a charging current of 0.5C (1.5A) and a cutoff at 150mA.

In the subsequent tests, I used a current cutoff of 150mA. I don't think it makes a critical difference in this case, but it's worth mentioning for completeness.


Since these batteries are more high-drain than high-capacity, I used a new ZKE-A20 electronic load for the test, with a maximum discharge current of 20 A. The load is new, so just to be safe I calibrated it before starting the test to improve the accuracy of the measurements.
I used the appropriate battery holder with a full four-wire connection.


Test conditions:
Ambient temperature: about 24–25 degrees
Cutoff voltage: 2.5 V, as specified.
Charged to 4.2 V at 1.5 A, with charging stopped when the current dropped to 150 mA.
After charging, I let the battery rest for about 20–30 minutes.

Ideally, I wanted to get a graph like the one shown in the datasheet.


Unlike the manufacturer, though, I plotted the graph at seven discharge currents rather than five (1C, 5 A, 10 A, 15 A, 20 A): 0.2C, 0.5C, 1C, 5 A, 10 A, 15 A, and 20 A.
Plotting the graph for one battery takes about (2 h 15 min × 7) + 5 h + 2 h + 1 h + 30 min + 20 min + 15 min + 10 min = 25 hours, not counting breaks.

First battery.
The color coding is very simple, and I think familiar to anyone who remembers the rhyme about every hunter wanting to know where the pheasant sits.
Red - 0.2C - 600 mA
Orange - 0.5C - 1.5 A
Yellow - 1C - 3 A
Green - 5 A
Light blue - 10 A
Blue - 15 A
Violet - 20 A



I don't have a temperature probe, so I had to measure the temperature with a thermal camera.
End of discharge at 10, 15, and 20 A, respectively. The manufacturer lists a maximum temperature of 99 degrees at 20 A; in practice, it was lower.


And the second sample. Unfortunately, I couldn't always leave the batteries to rest after charging. The manufacturer probably uses higher-end equipment; I don't have anything like that yet. Maybe they'll go on sale someday and I can at least buy a handle or a cable—I doubt I'll be able to afford anything more :)



I've run similar tests before, but this is the first time I've plotted the curves together. I also measured capacity by discharging to a specific voltage, such as 3.0 or 2.8 V. With a graph like this, you can also easily estimate capacity at different currents and discharge cutoff voltages.
For example, to find the capacity when discharging at 10 A down to 3 V, find the mark for the required voltage on the left, draw a horizontal line until it intersects the relevant curve, then drop a vertical line from that point. That gives you the capacity in that operating mode.
The scale marks are spaced fairly far apart, but the scale is linear, so I don't think it will be too difficult to estimate. In my example, the capacity is about 2550 mAh at 10 A and 3 V.


And that's the review. Here are my brief conclusions.
These are high-quality batteries that meet their stated specifications and even slightly exceed them. Their internal resistance is a little lower, which is probably why the temperature at 20 A was only 92 degrees rather than 99.
By the way, about the current rating: the batteries are rated for 15 A, but the datasheet also gives figures for 20 A. I assume they can handle that level briefly too—for example, in a cordless drill, where the average load isn't very high but there are short overloads.
I'd classify these as medium-current batteries, if that's a useful distinction, since it's hard to draw clear boundaries.
The closest equivalent is the LG HG2 “chocolate” cell; it has the same capacity and similar specifications.

Would I recommend buying them? Absolutely. It's just a shame that so few stores offer a near-100% chance of getting an original, while on AliExpress the odds of getting a clone or counterfeit are just as high.

That's all for today. I hope you found this useful. To be continued.