According to some old data I posted back in july 2013, when I was installing it, the temperature of the LED test point is 110 degrees with no chimney, and 80 degrees after the chimney was added.
Right now, I did a similar test with thermal imager.
Without a chimney, within 30 minutes the temperature of heatsink reaches 121°C. With the chimney on, the temperature after 30 minutes is 94 °C. Still getting this near-30°C-drop, but substantially higher absolute values (it seems there was something wrong with thermocouple-soldering way I did it previously. Or is it just dust?)
The drop may not sound like being much, but it can make a hell of a difference for LED lifetime.
If you are wondering, wtf is on the right side of the photo. It is.. no.. it was a halogen lamp that I used before this LED light. Until it blew up like a bomb, throwing red-hot pieces of quartz all over the place. Guys, don't use halogen lamps without shields. No, really.
Hi !
The passive cooling sytem is excellent :-) I've used it for a sophisticated project in 2012 and it was the only solution. It works great but the higher the chimney, the better efficiency. So I've used very long aluminium tubes (3m tall) and it saved the day !
The bottom however was another story, it took 10 prototypes to get right...
https://www.facebook.com/media/set/?set=a.257975450964048.59940.101308076630787