XG NCR21700 Batteries

I got my hands on another batch of batteries, and this time they’re 21700 cells rated at 4800 mAh.

A reminder: to stay up to date with new posts on the blog, I recommend following me on Instagram, where I’ll post notifications about every new article and maybe write about something interesting from time to time — link to my account

These batteries were given to me for testing, and also so I could get a better idea of what they’re actually like. I decided to put together a mini-review. I got them offline; the link in the title goes to visually identical cells on Alibaba. I also found Litokala batteries on AliExpress that seem to be made using the same cells — link.


There’s no marking that would identify the manufacturer, so I referred to them in my tests as XG NCR21700 4800mAh.
The only thing that sets these batteries apart is the three stripes on the negative terminal. You can see the same three stripes on batteries “for” or “from” Tesla car batteries.


There’s a barcode, but trying to read it didn’t yield anything useful beyond 123456789A. The code isn’t unique, so it’s not much use here.


Dimensions and weight.


During the “conditioning” run, I noticed a significant difference in capacity, which is a little strange.


For the tests, I charged the batteries at 0.5C to 4.2 V, terminating when the current dropped to 0.05 A.
Charging in this mode took almost three hours, which seems a bit long to me, especially given how drawn-out the CV phase was.


Almost all common cells of this capacity and size have a maximum discharge current of 3C. There is one exception, but that battery is rated at 5000 mAh, not 4800 mAh.
That’s why I tested them at discharge rates of 0.2, 0.5, 1, 2, and 3C, where C = 4.8 A.


Temperature at 1, 2, and 3C. I initially thought about testing at a higher current, but changed my mind after seeing the temperature at 14.4 A.


The second battery, tested at the same discharge rates.


The temperature is about the same.


A test of how closely the cells match. Sometimes this is even more important than simply measuring capacity.
Here, you can see that the second battery really does have a higher capacity. I wouldn’t call the difference critical, but I’ve come across much better-matched cells.
What was especially disappointing is that neither battery managed to deliver the rated 4800 mAh, even though charging was terminated at 0.05 A and discharge continued down to 2.5 V rather than 2.75 V, as I later found specified in one of the descriptions.
In other words, if charging is terminated at 0.1 A and discharge at 2.75 V, the capacity will be even lower.


Battery impedance as received, after “conditioning,” and at the end. The photo shows that the batteries arrived at different states of charge; they may have already been tested before they got to me.
Overall, these results are quite typical for batteries like these, so everything looks fine.


More as an experiment, I decided to compare the results with those for the closest equivalent batteries — Samsung INR21700-48X.
There’s a slight difference in both the results and the curve itself, but overall they’re more or less the same.


Conclusions.
I’d call these good batteries if it weren’t for one thing: based on the test results, I somehow got the impression that they’re used. Two things point to that:
1. Capacity: a bit low for a new battery.
2. Variation in the specs—or, more specifically, in capacity. I wouldn’t call it large, but it’s noticeable even between just two samples.

That said, the batteries passed all the tests without any problems, so they could be useful at a reasonable price.