Two different 21700 batteries give different output at all levels on C8F

I recently acquired a Sofirn C8F flashlight that runs off a 21700 battery, or a 18650 battery using the included sleeve.

Then I purchased a Samsung 50E 5000mAh 21700 battery as a backup for it. I knew that the 50E may not have the current capability of the included Sofirn 4000mAh battery at the turbo level but I was OK with that since I don't use the turbo level that often. Besides the next level down is just fine for me and I assumed that I'd be able to hit that no problem with the 50E.

So after getting the 50E in the mail today, I proceeded to charge it up along with the Sofirn battery so that I could test its performance. One after the other I tried both batteries in this flashlight and noted the Lux reading in my makeshift lightbox. Naturally, I expected to see a slightly lower reading with the 50E at the turbo level…………..which I did. And I expected to see the same output at the lower levels regardless of what battery was installed………….

But instead what I saw was a lower level with the 50E battery at all levels. At the low level (not the moon mode…..just level one), I'm getting 410 lux in my light box (about 7 lumens) with the 50E. And I'm getting 660 lux (about 11.8 lumens) with the Sofirn battery. It's a proportional difference like that at all levels.

So I'm wondering why that is? Evidently this light is not current regulated? I guess I can be OK with it as long as I get proportionally longer run time with the 50E compared to the Sofirn battery. What type of driver design would cause the output to be proportionally different at all levels with one battery vs another? :thinking:

my guess is that it is current regulated but not voltage considered. most amperage regulation also uses voltage monitoring but if this doesn't then the lower level amp output will limit the light.
or it just pain knows and is aware the battery isn't the one it wants but then they would tell you that. you could try running the test a few more times with a friend, having your friend change the batteries out so the test is blind and you have no predisposition of the result.

Thanks for the suggestion. I think I'm going to try with a 18650 battery to see if I get varying results too. I've got a Panasonic NCR18650GA and a Samsung INR18650-30Q. Both of those batteries can source a bit of current. At least the flashlight wouldn't be starved at the lower end. The only difference between batteries is their internal resistance governing the total capable amp draw. So here is a guess. I wish I knew more about this driver. I'm wondering if the light uses some type of PWM to the point that the battery is pretty much shorted into the LEDs for the duration of the on time of each PWM cycle. The length of PWM being fixed per its output level. Thus if that was what they are doing then a battery with a lower resistance would source more current during the on time of the PWM cycle and you'd see an overall average higher output with a lower resistance battery than one that is higher.

Anyhow that is the only explanation I can think of. If so I'm kind of disappointed. This flashlight uses 3 LEDs and I don't know if they are in series or parallel. If in series, I'd expect some type of boost circuit which I would think would be better regulated. So that makes me wonder if they are in parallel.