aracer
Hmmm……I do see where you are going with your effciency argument but it doesn't feel quite right. Below is extracted from a Luminus thermal application note. Apologies for poor formatting!
Consideration of Optical Power
In the previous examples, for simplification purposes, it was
assumed that all electrical power was converted to heat. Obviously,
this is not the case and a portion of the input electrical
power is converted to light. The amount that is converted into
optical power depends on the color of the LED and the drive
conditions of the LED. Since different color LEDs have different
efficiencies and the efficiency of LEDs degrades as a function
of input power, each use case will be different. For the most
up to date information on the optical power emitted from
PhlatLight devices, please consult the latest product data
sheets.
Once the optical power emitted from the LED is known, it can
be factored into the calculation of junction temperature. The
thermal resistance equation (2) can be modified as follows:
(4)
Thus when optical power is taken into account, the demands
on the heat sink are reduced.
Variable Red Green Blue Unit
Current 8.1 8.1 8.1 A
Voltage 2.3 4.3 4.1 V
Optical Power 2.3 1.8 3.5 W
Power Dissipated 16.3 33.0 29.7 W
Tamb 30 30 30 ºC
Tjmax 80 120 120 ºC
R? j-hs 1.32 1.32 1.32 ºC/W
?Tj-hs 21.6 43.6 39.2 ºC
So, in terms of efficiency when comparing input electrical power to output optical power using Watts, the above efficiencies range from about 5 to 15%. Now I know these figures are for colours but I believe a white LED is based on a blue device with a phosphor doping so its probably fairly representative?
What do you think?