I've always been curious about users getting higher lumens with higher volts vs Malkoff's claim to being regulated.
I've been curious about that as well. Running these dropins at higher voltage produces a flat output curve. The higher voltage, the flatter the curve. It makes sense, because the driver is regulating the output. However, the interesting bit is that
despite having crossed the regulation threshold, the output still increases with higher voltage.
So, correct me if I'm wrong--The driver controls the output by regulating
current draw, meaning that at a higher voltage, but drawing the same current, the emitter sees a slightly higher
wattage and the output increases. This explains why running the same dropin at 6, 8, 9 or 12 volts produces similarly flat output curves but at slightly different outputs. Right?
However, this is the weird bit: Looking at INFRNL's graphs, the same dropin run at different voltage levels appears to maintain its higher starting output for the duration of the test! Why is this? Would this mean that a M91T on 2x18650 (8v total) and a M91T on 3x18650 (partially depleted, 8v total) would produce the same output, even though a M91T on 3x18650 (full charge, 12v) would produce a brighter regulated output?
Another question this raises is: What would the graphs look like if the tests were split into three phases, separated by turning the light off and on during the test? Would we see three well-regulated segments at successively dimmer levels as the "starting voltage" dropped?
[Edit] I believe I've got some faulty premises here. My observations seem to apply to the M61T run at 4, 6 or 9 volts. Not so much the M91T. I still think the questions are valid.