Ok, someday I will do that, such a pita though.
I did a couple of tests on current draw at low on the Manker E01. 3 batteries, 3 different current draws. The voltage shown below is without load, I don’t have a convenient way right now to measure under load unfortunately.
fresh Duracell,V=1.61V, E01 draws 5.4mA on low, impedance 298 ohms
Fresh eneloop, V=1.41V, E01 draws 4.5mA on low, impedance 313 ohms
old kirkland, V=1.26V, E01 draws 3.6mA on low, impedance 350 ohms
old Duracell, V=1.22V, E01 draws 3.5mA on low, impedance 349 ohms
I am not sure if this is some sort of joule thief type circuit, some other PWM, or just an interesting boost that saves current as the battery voltage drops. Pretty easy to see how a 750mAHr eneloop that stores more than 800mA hr could last for 200 hours. If the roughly 320 ohm impedance holds, when the battery is in that 1.27V range the current is 4mA, or 200 hours worth of storage. Would love to see the circuit and understand how the driver works.
Repeated the test on medium level. I calculated impedance here but I think it is meaningless without knowing the actual battery voltage under these heavier loads, especially for the “softer” alkaline batteries.
fresh Duracell,V=1.61V, E01 draws 105mA on medium, impedance 15.3 ohms
Fresh eneloop, V=1.41V, E01 draws 097mA on medium, impedance 14.5 ohms
old kirkland, V=1.26V, E01 draws 077mA on medium, impedance 16.3 ohms
old Duracell, V=1.22V, E01 draws 077mA on medium, impedance 15.8 ohms
All in all, a fun way to spend part of a Monday evening. I think I can scrounge up another DVM, but in reality I learned what I wanted to learn, that is the current is as low on the low setting as I imagined, and it is not trying to regulate constant voltage/current on the LED on any setting, but looks more like a fixed impedance.