You can emit all the heat in the world, but it won't do you any good (in this particular application) if you can't conduct it off an emitter or a heat sink
That's true, of course, and it's why a poor conductor of heat poses problems for higher powered LED lights. You have the junction between the LED die and the heat sink, and the junction between the heat sink and the light body, both of which require good conductance. So, I agree.
But . . . the junction between the light body and the air (or hand) has a lot to do with both conductive and radiative heat transfer (and also convection, I suppose). And bare Ti beats bare Al here. However, HA Al beats Ti, if I remember correctly.
The real question to ask about a Ti light is if it has adequate heat sinking and heat transfer to keep the LED die temperature low enough for good light output and longevity. Consider my favorite light, Titanium or otherwise: the
LunaSol 20. Don did both a runtime/output test of the LS20, and a thermal test of the LS20 towards the end of that thread (post #104). I stole his images to post here, and also because I will be using them (unless he objects) in my LS20 review.
First the thermal setup he used:
The purple masking tape has an emissivity of 1 and allows for an accurate thermal image. The DVM is measuring the temperature of the copper penny, which allowed Don to accurately determine the surface temperatures of the light. And here is the thermal image:
The hottest part of the lights surface is 47.7 C, which is just about 118 F. This is with the light sitting out in the open after 15 minutes of continuous running on high. This is a worst case scenario. If the light is held in the hand, it doesn't get even that hot when running continuously on high. Now, the die temperature is higher than the surface temperature, but even with only very conservative values for the thermal resistances of the junctions, the die temp is well within limits. And another way to tell this is to look at the lumens output over time. If the die temperature gets too hot, lumens output falls off. So if you see an output graph of a light which starts high and drops more or less linearly to a constant output level until battery depletion, you are seeing the effect of a rising die temperature, rising too high. I remember seeing this on one of the Arc lights. The Arc4 prototype, IIRC, on high. this_is_nascar commented on it, and Gransee told him that if he just held onto the light the entire time, the output would not drop off. TIN doubted it, tried it, and sure enough, Peter was right. Anyway the point is that if you
do not see any kind of output drop off in a current regulated LED light tested while just sitting in open air, then the thermal design is more than adequate to the demands placed upon it.
Here is the output graph of the LS20 on high:
As you can see, there is almost zero drop off.
So, we can argue about numbers and theories and yield strength all day long, but ultimately, the practical results and tests speak for themselves: Titanium has good enough thermal characteristics to be used in the body of an LED light.
Period.
Should it be used? Is it the
best choice in all situations? Is it for you? These are questions only you can answer. Your money, your choice. But, let's please stop saying or implying that Titanium is unsuitable because it doesn't conduct heat well enough or doesn't have a low enough electrical resistance. It handles both current and heat just fine, thank you very much!
