So in the interests of science, I have conducted another test.
It's now just after 10pm here. This is an
outdoor tailstanding test. Light is sitting on a chair in the driveway with a decent airflow across it.
Ambient temperature is 18C and there is a gusty breeze blowing. Again, I have no IR equipment so results are subjective.
Same Panasonic cell as earlier today - recharged and with a resting voltage of 4.16V
At 3 min mark the head is warm and the body somewhat cooler At 5 min mark it is definitely warmer but not uncomfortable to touch in any way. Indicator light still blinking normally
At 10 min mark, I can still grip the head comfortably though it is now what I would call quite warm. Indicator light still normal
At 15 mins, no obvious change that I could detect. Body of light has increased in temperature and is no doubt helping to shed heat
At 30 mins I decided to stop the test as we seemed to have reached equilibrium and I didn't see the need to run any longer. Indicator LED still normal slow green flash.
Resting battery voltage 3.65V
My conclusions:
The heatsinking on the headlamp seems to be very well-done considering size and output. Remember, we are not dealing with a TN31 here but a tiny light that fits into the palm of your hand. The fact that the whole body warms as the light heats up indicates to me that the thermal path is well-engineered and performing its job properly.
I'll be interested in re-testing during summer when the ambient night temperatures can sometimes exceed 30C but for the moment I am very satisfied that this thermal stepdown "problem" won't be an issue for me in daily use.
In the event of a blackout, I would consider that the 550lm mode would provide more than enough light for ceiling bounce and others have reported no stepdown issues at this level.
Make of these results what you will.