I have the same effect with my 4000K Version. The thermal setting are configured correctly, room temperature is reading accurately. Temperature limit is set to 60° (Thermal Cfg -> 2. +30 clicks).
On Turbo it works fine, the host reaches ~60° and the regulation begin slowly to step down.
But in ramping mode, on any ramp level about 1000 - 2000 lm, the regulation kicks in after 10 - 15 sec. and goes rapidly down to 500 lm. No matter what temperature the host have, it reaches not even 30° and nevertheless it begin to step down. After some time the output level goes upwards, but the host temperature does not reach the set temperature limit at all, I did not measure it, but about the 45 ° max. None of my other Anduril lights behave like that.
Does anyone have an idea why the thermal regulation behaves like this? Seems like a software problem.
Would be nice if someone else would test the MF01 mini on this behavior.
Fuck me. I’ve just noticed mine is doing the exactly same thing, I also have the sst20 4000k version. Its regulating down on max ramp (not turbo) after few seconds of use. Its not even hot to the touch jet, just a bit warmer than environment. And it drops to much lower level. It stays much longer on turbo than max ramp. I have those copper and brass on the way… How to fix this? Is it something in andurils configuration for this light? Does anyone know?
freeme? Is this bug in all astrolix mf01 minis? Should we flash the firmware? Because it very limits the usefulness of the light to have it termo regulate down so quickly… Is this a firmware problem or is it caused by some hardware component? Can anytging be done about it?
Since we have 2 software related questions, stepdown and lvp, I sent a PM to Toykeeper with this thread link. Hopefully she will post here or answer me and I’ll relay here. Now we wait.
I don’t want to unscrew the flashlight again so I will give you the dimensions of glass instead inner diameter of bezel (all dimensions are approximately ± 0,2mm)
Glass - 40mm x 2mm
MCPCB - 32mm x 2mm
Optic - 34,75-34,85mm x 9mm
Thank you very much, Chatika vas Paus. These are exactly the data I have been looking for. Your fotos are excellent and give a good overview of the internals.
I remeasured using a high drain Golisi 4300mah cell and got 3,860 lumens at turn on compared to 3,770 with the Vapcell 26650. Meaning the SST-20 4000K is pretty much pushed to the max already even with the Vapcell 5500mah.
You think it’s maxed due to the led forward voltage and not anything related to the driver or battery circuit limiting output? This distinction is something I’m trying to figure out. It’s a bit confusing.
PS, it’s also good to record beyond 1s, like to 10 and 20 seconds for instance just to see how quickly and how low the output drops due to the battery load and leds heating up.
I found out that the LVP goes by the same voltage it blinks out (the MCU blinks the voltage it sees and LVP is controlled by the MCU). So if the voltage reads a bit lower than actual, LVP will activate a bit sooner.
I dont think there is much in the way of fixing the thermal stepdown behavior. On some driver designs the stepdown works well and on others it acts not so great.
Has anyone tried setting the top of the ramp a tiny bit higher or lower and then checked how the thermal stepdown responds? Maybe it will change.
and bad health 
These are exactly the data I have been looking for. Your fotos are excellent and give a good overview of the internals.
