Turned all on MAX, and the Convoys’ heads got hot really quick, while F8 was heating up very slowly.
So it looks like F8 does a rather poor job dispatching heat from LED.
Why is that? There’s not separate pill, but is an integral part of the body. Mayby the MCPCB is not pressed hard enaugh by the reflector? Or maybe the body part that MCPCB sits on shout NOT be anodised (but it is).
The “pill” anodising looks like this (it’s not F8, but it looks almost the same):
Compare 8 vs 8 chips not 8 vs 6 chips, with a luxmeter you could see how the brightness drops compared to each other and relate that to the temperature.
My bad that I didn’t check the light output when performing the test.
The output gets lower when the temperature goes up, so I would be able to say whether the heat is dispatched from the emitter or is it overheating.
remove the ano from the lip, make holes for screws so they will press the led star tightly and dab some thermal conductive paste on the lip - should work much better
The pills on the Convoys are also much smaller and shallower than the more spacious pills on the F8. The design of the heads and bodies are also different in thickness and design, so there really isn’t much to base a comparison on.
That’s right, but… Roche F8 and Convoy M2 have the same LED and the same driver. So, when on MAX emitters generate the same amount of heat. Obviously this heat must be moved outside through the flashlight body. The M2 doest that, and the question is - regardless of body and pill design - what’s happening to the heat inside F8?
Janko - I’ll try your solution, as I am almost sure, that the heat cannot cross a barrier between the led star and the “pill”.
anodizing blocks heat transfer a bit… and without any thermal conductive paste the transfer doesn’t work well because of the irregularity of surfaces - they should be as smooth and straight as possible and the paste will fill the tiny irregularities where air would be without it (and you know that air is an insulator)
the screws would only help to press the star tight and secure
Convoy S3 is a tiny light with considerably less surface area compared to a F8.
Lets see a 15 minute test comparing M2 vs F8. Followed by a 10 minute cooldown (with measurements). Add S3 if you wish.
Its not like a flashlight can “hide” extreme amounts of heat in the MCPCB.
I would suggest the star is not properly attached to the host iF the heat is really not getting out, the f8 is almost identical to the cnqg d5 I built, running at 3.1a, it soon starts to warm up on high. Dont forget, there is no pill, so no pill - head junction to over come, the heat should transfer directly into that nice heavy lump of aluminium with minimal fuss.
In comparison, the m2 and s3 are both a poor design in that, heat must transfer from star to pill and then from pill to head.
The d5 is also considerably heavier than both, more mass means more alu to heat up before you start to feel it.
Unfortunately it is very hard to measure the LED itself (which is what we care about) so we have to judge the thermal performance by measuring the body temperature, and consider this problem as a conservation of energy.
The biggest downfall with measuring the body temperature is that we would need to measure the temp. difference of the entire body, not just the max. I suspect that the entire head is heating up on the F8, so the energy being put into the head is spread out over a lot more aluminum (through conduction), and showing a smaller max temp (*assuming a good thermal path from the LED to the body). The same energy placed into the head of an S3 would result in a higher max temperature and a much faster temperature change like your chart shows.
A second part of the problem is the energy lost to the environment. The greater the surface area on the body of the flashlight, the more heat energy can be released into the surrounding air through convection. That’s why we have lots of fins on computer chips with a fan nearby.
Taking these two together, you can say (in general) that a higher thermal mass (more material) will result in a slower maximum temperature change, and a greater surface area will result in a lower maximum steady state temperature.
