Molicel INR21700-P42A: More Batteries from E-One Moli Energy

This time, we’re looking at high-current Molicel batteries again, but in the newer 21700 form factor. Well, not exactly new: batteries of this size have been around for quite a while, but they’re still much less common than the more widely used 18650 cells.
In any case, I liked these batteries, but you’ll have to read the full review to find out more.


First, the price. The title lists a price of €7.95, but if you buy more than 50, it drops to €6.85. Unfortunately, this option doesn’t work for residents of Ukraine, since the total would then exceed €300, while the limit is only €100.

Now, about the batteries themselves. When I downloaded the datasheet, I found out that these are fairly new batteries: datasheet version 0.2 is dated August 2018. So it seems there aren’t any batteries with an older production date yet.


As before, I bought a few batteries, two of which went into testing.


According to the description, the batteries have a minimum capacity of 4000 mAh, but the model name ends in 42, which looks more like an indication of 4200 mAh capacity. Also, the charging current is 4.2 A, which indirectly suggests a capacity of 4200 mAh rather than 4000 mAh.
The maximum discharge current is 45 A, but it doesn’t say whether that’s continuous or short-term.


As with the batteries in the previous review, the description includes all the necessary information about dimensions, weight, and how to decode the production date.


According to the code 2IB23, the batteries were made on November 23, 2018. There’s also a 2D code, similar to a Data Matrix code. Besides providing useful information, it also helps prevent counterfeiting, since it’s unique to each individual battery.
The positive terminal protrudes slightly beyond the casing.


As last time, there’s a drawing of the positive terminal with its dimensions.


The situation with 21700 batteries is still a little puzzling. They often have decent specs, not much higher than those of the popular 18650s, and let you build 1P battery packs with higher output current. But somehow, they still haven’t become widely adopted.


My charger, which usually overestimates capacity by quite a bit, showed only slightly more than 4200 mAh in this case. But more detailed tests will show “what’s what.”


For further testing, I used an EBC-A20 electronic load with a maximum load current of 20 A. The battery was installed in a heavy-duty holder with a four-wire connection.


As I mentioned above, the description specifies a charging current of 4.2 A, but also says the maximum current can be as high as 8.4 A. There’s a small graph further down that explains the conditions under which the batteries can be charged at that current.
According to the description, charging at 8.4 A is allowed while the battery casing temperature is between 15 and 45 degrees. In all other cases, the current must not exceed 4.2 A.


Although the temperature conditions in my test were suitable for charging at the higher current, I’ll show the graph for 4.2 A only. I wanted to try charging at 8.4 A, but then remembered that my charger has a maximum current of just 5 A. I tried that too, but the difference was very small.
A full charge takes 1 hour 15 minutes, which is very good for a charging current of 1C.


Curiously, unlike the previous batteries, the description lists only three key load-current values, with the maximum being just 10 A, even though the previous model was rated for 20 A. To be honest, I didn’t quite understand that.


But let’s move on to the tests. They were performed at the following load currents:
0.2C (840 mA) - red
0.5C (2.1 A) - orange
1C (4.2 A) - yellow
10 A - green
15 A - light blue
20 A - blue.

The 2C (8.4 A) test wasn’t performed because it’s very close to 10 A.

The result was a baseline capacity (at 0.2C) of 4060 mAh (14.86 Wh), while at 10 A it was 3954 mAh, which is fully in line with the figures given above. At 20 A, the capacity was only slightly lower, at 3932 mAh or 13.33 Wh.


Temperature was also fine: at a discharge current of 20 A, the battery warmed up to just 72–73 degrees. Below are the thermal images at 10, 15, and 20 A.


The second battery delivered about 20 mAh more, but overall its results were much the same as the first one's.


The temperature was slightly higher, around 75 degrees at 20 A.


As you can imagine, I didn't stop there. Until I get an EBC-A40, I have to make do with whatever I have on hand. In this case, I used a load with a maximum current of 30 A.
The procedure was the same as before: voltage was measured directly at the battery, and I adjusted the load settings to compensate for the voltage drop across the power cables. The cutoff was set to 2.5 V.


In the end, the first battery delivered 3922 mAh, and its final temperature was 98.8 degrees.


The second delivered almost as much, 3920 mAh, and reached 98.7 degrees.


Although this load can be connected to a computer, its Chinese software refuses to work, so I connected a multimeter in logging mode instead.


The voltage before the test was 4.185 V. The batteries had been sitting for several hours after a full charge.
The discharge cut off at 2.51 V.
The average voltage was 3.288 V.

The voltage dropped slightly right after the test began, but then started to rise—you can even see this on the graph. After that, it fell as the battery discharged.


After letting the battery cool completely to room temperature, I was able to discharge another 81 mAh at 0.2C.


Next, I tried discharging the battery at an even higher current, using a load rated for up to 40 A.
The test procedure was a little different this time. As before, I let the batteries rest for about 8 hours after a full charge, but I stopped the discharge manually. There was no need to go into a menu to stop it; I could just press a button.


Unfortunately, because of the load's limitations (it's a high-voltage model), the maximum current I could get was 36.4 A. As the battery discharged, it dropped to 27.4 A.


The first battery delivered 3945 mAh and heated up to 102 degrees. The room temperature was around 20–21 degrees.


The second battery delivered almost as much, 3940 mAh, but got slightly less hot, reaching 100 degrees.
I ran a similar test on Samsung INR21700-40T batteries and had to stop the discharge because they were overheating. However, the room temperature was about 6–7 degrees higher then.


I also plotted a discharge graph for the second battery.


The battery's internal resistance is about 9.5 mΩ, and it increased slightly after all the tests.
The photo shows the results when I received it, after “conditioning,” and after all the tests.


For now, the only battery I can compare this one with is Samsung INR21700-40T, which has roughly the same rated specifications and size. I did review a similar battery from Litokala here, but back then I couldn't make graphs, and it performed worse in the 30 A test.

In any case, the Molicel battery performed better than the Samsung INR21700-40T. The difference isn't large, but it's there.


To sum up the test results, I liked these batteries. Unfortunately, although they're a little better than their closest competitor, the Samsung INR21700-40T, the price spoils the picture: Molicel batteries cost about one and a half times as much. Maybe they're more reliable, maybe there's some other advantage—I won't argue the point. But right now, the Samsung INR21700-40T is simply the better deal.
The only argument in Molicel's favor is that they're rated for currents up to 45 A, while the Samsung is rated for only 35 A. But in my opinion, they perform exactly the same.

That's all. As usual, I'd be glad to hear your comments, and I hope you found this useful.