Basen BS26003 batteries, 4.5 Ah in the 26650 size

Another pair of Basen batteries, but unlike the more familiar 18650 format, these are larger, 26650, and accordingly have a higher capacity: 4.5 Ah.
Inspection, tests, conclusions.


This battery format is used much less often, possibly because of its less favorable capacity-to-size ratio, or perhaps simply because it is large: too big for portable devices, but too small for stationary ones. I don't know, but the selection is clearly smaller.

Unfortunately, by the time I finished writing this review, the discount had ended and the price had gone up from $5.75 to $6.39 per battery. That makes them less attractive, and a pair isn't much cheaper either. What a shame...

This time there are no nice plastic boxes, just cardboard packaging.


I did find the technical specifications on the Basen website. They're very brief, of course, but still. It's a little surprising that the stated charging time is 470 minutes at 1.2 A, but only 410 minutes at 4 A. To me, those numbers look like they were just made up.
Of course, you could say that perhaps it uses a time-based charging algorithm, like Murata batteries, for example. But the charge termination current is specified here too, so it's clearly not charging by time.


For protection during shipping, the batteries are wrapped in clear heat-shrink tubing, which needs to be removed.
The capacity, size, and peak current of 60 A are printed on the batteries themselves. The positive terminal protrudes noticeably from the end of the case; the negative terminal does too, but much less.


When I started looking for equivalents, I wasn't even surprised that the closest match was, once again, VapCell. Thanks to the well-known Dane's website.
The technical specifications were there too.
Product: Vacell 26650 4600mAh
Model: IMR 26650
Size: 26.2 * 65.2 mm
Nominal voltage: 3.6 V
Charging voltage: 4.2 V
Maximum charging voltage: 4.25 V
Typical capacity: 4600 mAh (0.2C discharge)
Minimum capacity: 4500 mAh (0.2C discharge)
Top: flat top
Color: yellow
Weight: 98 g max
Internal resistance: 3 months: -20 °C ~ 45 °C

There is a small difference: VapCell specifies a continuous discharge current of 40 A, while Basen specifies only 30 A. Also, VapCell's standard charging current is 1.5 A, compared with 1.25 A for Basen.


The listed weight is 86.9 g ±0.5 g, but in reality they weigh 90 g.


On the test run, the charger delivered almost 4800 mAh, so there's a chance the actual capacity is even higher than the stated 4500 mAh. However, the discharge current was 500 mA, which is almost 0.1C.


When preparing the tests, I initially used the AliExpress product page as a reference, but it has no technical data. So I chose a charging current of 1.5 A based on information I found on the Danish website—that is, the VapCell battery specifications.
At this setting, a full charge takes almost 3.5 hours. That's a very long time to charge such a high capacity at such a low current.


A charging current of 4 A is also allowed, and this is the fast-charge mode for these batteries. A full charge took almost 1 hour 40 minutes in this mode, but I used 1.5 A for the subsequent tests.


The tests were performed using a ZKE EBC-A20 electronic load and a compatible battery holder with a four-wire connection.

The discharge tests were run at 0.2C, 0.5C, 1C, 2C, 15 A, and 20 A. Here, C = 4.5 A.

The battery has an unusual discharge curve: at currents of 15–20 A, the capacity delivered in mAh starts to increase rather than decrease. This is due to the battery heating itself up, something I've seen with Murata VTC4 batteries and which is also noted in the datasheet.


The temperature rise is fairly moderate, but even at this point I can say these are medium-current batteries—in other words, all-purpose batteries, not high-current ones as claimed.


The second battery showed roughly the same results. The only difference is that its capacity started increasing at a current of 9 A.


The temperature at the maximum current was a little lower, though. I'm writing this review now and wondering if I might have missed the end of the discharge, even though I always try to keep an eye on it. The difference is rather large. In any case, it is probably more accurate to take the maximum as 87 °C, which is what the first sample reached.


Someone once wrote to me that a battery can't deliver more capacity in mAh as the discharge current increases. But the thing is, the Dane got exactly the same results with his VapCell, and since this kind of behavior is more often the exception, we can say we’ve found another similarity.


The batteries are rated for a maximum continuous discharge current of 30 A. I was skeptical, of course, but decided to test it anyway.


The test method is simple and straightforward. I connect the holder’s power leads to the electronic load and the sense leads to a multimeter. I install the battery and start the test, watching the difference between the readings on the load’s voltmeter (it has a two-wire connection) and the multimeter as the test begins.
Then I turn off the discharge current and adjust the cutoff voltage. In this case, for example, the difference was about 0.55 V, the voltage drop across the holder, wires, terminals, and so on. So I set the load to 1.95 V (2.50-0.55=1.95).
Then I start the test again.


The battery started heating up fairly quickly. When its temperature reached 95 degrees, I stopped the test; by then it had delivered about 3 Ah.
After the battery cooled to about 70-75 degrees, I resumed the test. By the time it finished, the temperature had risen to 97 degrees again, but this time I simply waited for the load to shut off by itself. The total was 4.31 Ah.
Just for the sake of the experiment, I ran the test again, and got another 100 mAh or so.


In his review, the Dane complained about the severe voltage sag under high current, and I completely agree. Even right at the start of the test, the battery voltage dropped to 3.07 V. Then self-heating kicked in and it started to rise a little, reaching 3.1 V, before gradually falling as the battery discharged. The battery temperature was 68-70 degrees.


The second battery produced roughly the same results, both in capacity and temperature.


The experiment didn’t leave the batteries unscathed: the heat-shrink tubing started to creep. On the first one, it came off only at the negative end; on the second, it also split at the positive end.
VapeCell actually fared even worse, though either the temperature was higher or its heat-shrink tubing was more delicate.


Judging by the internal resistance, the batteries weren’t significantly damaged. Below are the results from before testing, after conditioning, and after all the tests. In total, each battery went through 8 full charge/discharge cycles at different currents.


And, as always, the conclusions.
First, a word about the cell manufacturers. I’m almost 100% sure these aren’t top-tier brands, if only because their 26650 ranges are very limited. Also, for example, VapCell shows up on Alibaba’s website far too often.


That said, the batteries performed reasonably well as medium-drain cells, with a maximum continuous discharge current of 15-20 A. That’s their actual continuous limit; I think they can handle close to 40 A briefly, but the severe voltage sag can cause problems, so it’s still better to cap the current at 25-30 A.
Why do I classify them as medium-drain cells? Cells of the same form factor tend to trade capacity for current (unless they’re ultra-budget models). For example, the maximum capacity of an 18650 is around 3500 mAh, while medium-drain models top out at 3000 mAh. For 26650 cells, the maximum capacity is around 5000 mAh, and a medium-drain cell can easily top out at around 4000-4500 mAh.

So I think these are decent cells, but not from a major brand, and clearly not high-drain. The battery is obviously being pushed too hard at high currents, although its discharge curve looks good at low currents.

As for long-term reliability, I can’t say anything definitive, and I doubt anyone else can either. You’d need to run life-cycle tests, which are time-consuming, expensive, and inconvenient. I really hope I’ll eventually be able to get my hands on a ZKE EBC-X load, which is designed for exactly these kinds of tests.


That’s all from me. I hope you found this useful.