The two cells match fairly well and maintain fairly stable capacity even at high loads, but the cell get hot at 30A (My holder melted).
Conclusion
The cells looks good, but I sort of doubts the continuous discharge rate of 45A, it will probably always be stopped on the 80°C cut-off before the cell is empty.
Notes and links
The batteries was supplied by a reader for review.
Nice, thanks! I snagged a few of these for only 10 CNY (1.50 USD) each on Taobao. They are like new but from disassembled battery packs. They seem to be very similar to Samsung 40T which is the gold standard for high drain 21700.
You wrote 45A several times in the review but I assume you mean 35A continuous discharge? I believe they claim 45A only with adequate cooling.
Oh my mistake, the datasheet does say 45A. Although the Taobao seller I bought it from rated it at 35A and said that active airflow is necessary for 45A.
It’s also confusing because Lishen gave the cycle rating at 35A.
By the way, there is probably a typo in the second image then where it says Rated A: 35.
The cell did well at 20A. The hottest cell of the pair reached a hair above +45°C.
At 30A Henrik's holder had a problem. I downloaded the graph, and using an image editor confirmed where the hottest cell would have landed if Henrik's holder were have been up to the task: +53°C. Considering that HKJ's setup presumes 25°C of ambient temperature, this is about the formal limit for li-ion cells.
So, 45A continuous? Nonsense, I'd say. Well, maybe if you stick it in a refrigerator, or use very good cooling. The average output voltage, though, would be rather low, with output voltage plummeting below 3.4V nearly right after the start of the discharge.
Tad off topic maybe, but enough already with the discharge rate thing, by the way. Energy density hasn't seen an improvement for ages by now. Sucks. And that's not to say that I also don't like a lot the sloping shape of the voltage curves for li-ion cells; stable output voltage, à la LiFePO4, is better. We already have super accurate algorithms for estimating state of charge, like coulomb counting and etc., and so we can do with a lot less slope in output voltage curves, which was the reason for batteries with continually decreasing output voltage in the first place. A new battery technology? Perhaps.
Is there something you don't understand in my comment above? I already know the answer, so here's an explanation:
The temperature graphs which Henrik publishes show the temperature above ambient which the cells reach. He substracts the ambient temperature from his temperature measurements to get temperature above ambient figures.
For this reason, to obtain the maximum cell temperature you must add the ambient temperature to it, 25°C usually.
The figures I posted above, +45°C and +53°C, are temperature above ambient figures. See the + addition sign just after the numbers? It's there for a reason. If we add the usual 25°C ambient temperature to them, we get 70°C and 78°C peak discharge temperatures. What does this tells you know?
Temperature above ambient figures lets you know more easily how a cell would behave in cooler or warmer conditions, particularly the latter. If ambient temperature were to be quite hot, like 40°C, and we add to it the aforementioned temperature above ambient numbers (+45°C at 20A and +53°C at 30A), we get 85°C and 93°C. This means that in such conditions even discharging at 20A continuous would require a bit of cooling to avoid the cells surpassing 80°C.
The initial voltage sag 40T obviously beats LR2170LA, it is the reason most people need 40T for. If you don’t care the initial voltage sag, P42A actually beats both.