Dunno, it's actually measured off a full up flashlight at the tail using either a wight v009 FET+1 driver or a MTN-17DDm driver, on both a Fluke DMM and a UNI-T UT50B DMM. No watchdog running, just the e-switch will wake it up. No idea if this is the right way to measure it or not. Again, I'm no expert at this stuff at all and don't claim to be. All I did was set the meter to read DC milliamps, moved the DMM lead to the connector marked for mA's, so I think I did it right?
If my measuring method is somewhat correct (no idea), then at least it proves our custom boards and firmware draw less amps than say a stock Eagle Eye X6R.
I can also see it step down when the power saving mode kicks in. I delay it for 10 seconds after turning the light OFF (PWM=0). For example, the ZY-T11 clone with no indicator LED was at 4.91 mA with the MCU running, then dropped to 0.302 mA after 10 secs (sleep mode).
Edit: did A few more tests of stock Olights, SWM, and NiteCore and it's interesting. These are much lower in power saving modes:
Olight S15: 0.001 mA, Olight S1: 0.002 mA
NiteCore MH20: 0.027 mA (button LED blink off), 0.061 mA (button LED blinking enabled)
So, there is something goin on here - why these lights are so much lower in standby than our designs. I thought we can't do any better in firmware (sleep mode with only the e-switch configured to wake it up), so would the 2 voltage divider resistors be actively draining power from the cell?
Edit #2: Again, not sure bout this, but using V = I * R, for 4.2v I calculate the 4.7K and 22K resistors in series would eat 0.157 mA. If so, that's bout 1/2 the current I'm measuring. Partially explains the amps draw at least.