[ QUOTE ]
Sign me up for a YYAR K2 LuxeonV!
[/ QUOTE ]
YYAR! There be a bright luxeon, else I be a crusty barnicle!
sorry, I just found that funny.
If lumileds keeps the luxeon V (4-die) line alive, it would indeed be impressive with K2 dies. I would suspect thermal resistance would be in the 4-6C/W range. If it were allowed to be run at full power (1500mA/die), then you're looking at a 3A, 7.6V Luxeon V - for 23W of power dissipation.
The binned rating for such a beast would probably be in the 600lm range. But, heat would take its toll.
At 6C/W (the most likely thermal resistance), Tj would initially be 163C, for a 30% loss in output, or a realistic output of 420 lm. You'd quickly hit the 185C max Tj with only a 20C increase in heat sink temp. This is just the J-Slug resistance. With an actual installation, that would be more like 7C/W, making things much worse
So at full power, binned efficacy is 26 lm/W, and a real-world efficacy is 18 lm/W. That's no better than a high pressure xenon/halogen bulb. You'd be better off running it at half power.
At half power (1400mA), binned output would be around 400 lm. Power dissipation would be approx 1.4A * 7.1 - about 10W. Initial Tj is 85C, real output is around 350 lm. Binned efficacy: 40 lm/W, real efficacy: 35 lm/W
10W and 85C is a much more realistic target for such a design. Any more power, and you lose efficiency very rapidly, and have a hard time dissipating that much heat in a hand-held light.
A quad K2 in a K2-sized package running at full power presents a daunting thermal management problem - there's just too much heat and not enough metal to trasfer that heat away from the junction. The should really go with a larger package that offers better thermal management if they're going to get into the double-digits for power dissipation in an LED.