Liitokala 32700 LiFePO4 and a comparison with similar VariCore batteries

Not long ago, I published a review of VariCore 32700 LiFePO4 batteries. At the end, I mentioned that I’d ordered the same batteries from the same store, but under the Liitokala brand. The batteries have now arrived and been tested, and, most importantly, I was able to compare them with the previous ones. It turns out there’s a significant difference. But as usual, you’ll have to read the review to find out all about it.


The surprises started right from the get-go. As last time, I ordered a lot of 10, which cost me $28.56 during an AliExpress sale. But whereas last time I received the batteries in a box with the Liitokala logo, this time they came in some unbranded packaging. As before, though, each battery was individually bagged.


The quantity checks out.


The product page lists the same specifications, so the store initially presents them as identical to the VariCore batteries.


In reality, there are only two differences: the label and the marking on the negative terminal. The previous batteries had markings like H6, H7, and so on; these all have a marking of 0.


The weight and dimensions are also identical.


Like the previous ones, these also have welded tabs. I’d probably have ordered them without tabs, especially since I didn’t like them last time, but the tabless version cost more, so this time I bought the ones with tabs just to save money.

This time, out of curiosity, I decided to measure the tabs. They’re 8 × 0.1 mm, giving a cross-sectional area of 0.8 mm². A section 18 mm long has a resistance of about 2.5 mΩ.
I’ve already reviewed nickel strip (I highly recommend this store), and included resistance figures there. For nickel, one meter of strip with a cross-sectional area of 1 mm² has a resistance of 54–55 mΩ. Here’s what I got:
2.5 × 0.8 = 2.0 mΩ for 18 mm of strip with a cross-sectional area of 1.0 mm²
1000 / 18 × 2 = 110 mΩ per meter of strip with a cross-sectional area of 1.0 mm²

The calculations show that this isn’t nickel, which should have half the resistance at this cross-sectional area. This was confirmed by load testing: the tabs heated up noticeably at currents above 15 A. Most likely, they’re made of ordinary tinplate (soft, by the way) with a nickel coating, since they’re quite magnetic.



Before confirming receipt, I tested all the batteries. It took a lot of time, though, because each one had to go through a charge-discharge-charge cycle.
The initial test showed that:
1. All the batteries were completely discharged—the first charge amount is shown in row 1-1.
2. All the batteries have not only the same capacity, but also a noticeably higher capacity than the VariCore batteries (row 1-3).


The VariCore batteries’ resistance varied between 8 and 12 mΩ, while all of these are around 6 mΩ. So it’s not only lower, but, more importantly, consistent across all the batteries.
The photos show the test results in pairs: as received and after the first full charge-discharge-charge cycle. Rather than squinting at the photos or making a table, I’ll just say the lowest reading was 5.86 mΩ and the highest was 6.4 mΩ.


As last time, I checked the charging speed in two modes, at 3 A and 5 A.
First, at 3 A: charging took 2 hours 32 minutes. The VariCore batteries took 2 hours 18 minutes, reflecting the difference in capacity.


At 5 A, charging took 1 hour 33 minutes, compared with 1 hour 18 minutes for the VariCore batteries.


The tests were performed with the following settings:
1. Charge termination voltage: 3.65 V
2. Charging current: 4 A
3. Charge termination current: 0.1 A
4. Discharge cutoff voltage - 2.0 V

At 0.2C, the resulting capacity was 6600 mAh, or 21 Wh, which is much higher than the VariCore's.


I didn't take a photo of the first battery heating up at 10 A. At 15 and 20 A, the temperatures were 57 and 64 degrees, respectively.


The second battery showed similar results


Temperature at 10, 15, and 20 A (the max reading is at the bottom of the photo)


Here's another look at a characteristic of LiFePO4 batteries: a test at 20 A, with a total runtime of 19 minutes, 7 of which saw almost no change in voltage.


The next test was run at 30 A, with the same charging and discharge cutoff parameters as before.
First, I connected the battery to the load and multimeter, measured the difference between the load's reading and the actual value, then adjusted the cutoff accordingly.


At the end, I got 6478 and 6461 mAh at a temperature of about 80 degrees.


Unlike the VariCore, the voltage rise as the battery heats up is slightly smaller here.


After letting the battery cool to room temperature, I continued discharging it at 0.2C (1.3 A), which yielded another 68 mAh.


The 30 A test had no effect on internal resistance. After the first cycle, the readings were 6.14 and 6.09 mΩ; at the end, they were 6.16 and 6.10, respectively, which is within the instrument's measurement error.


And of course, here's a comparison with the VariCore on one graph.
It shows the results in three discharge modes: 0.2C, 1C (6 and 6.5 A), and 20 A. I think the difference is obvious.


Based on the tests, I have to give these batteries a thumbs-up. The difference from the previous ones is enormous, both in the measured capacity and in how closely matched these batteries are. Now I really regret not opening a dispute over the previous ones. I also regret not ordering more, because now I'm left with 2+2 batteries that I can't use for anything :(

But, as always, there's one "but." Someone who bought Liitokala batteries earlier reported noticeably worse results, and several batteries had extremely high internal resistance, which points to inconsistent quality and batch-to-batch variation. Unfortunately, that's the curse of no-name batteries. Branded ones don't have this problem, but they cost noticeably more. So either buy a brand-name battery, such as A123, or buy Chinese ones and allow for some rejects.

That's all from me. I've tested and shown what I could, and as always, I'm happy to answer questions and hear your suggestions.