# Calculating of the heat dissipation of a LED

• NJZFW
In summary, the LED will generate more heat than it will consume, so you will need to design the enclosure and cooling for 100% heat generation.f

#### NJZFW

Hello Forum,

I write you with the challenge of calculation the heat dissipation of a LED.

The LED in question is http://www.seoulsemicon.com/en/product/spec/SZ5-M3-W0-00/22

The LED binning is 290lm (W4, 5000K) at 700mA, the efficacy is 157lm/Watt,electrical

I think I need to convert der chart "relative radiant Power over wavelength [%]" in "spectral radiant Power (W/nm) [%]".
But Iam not sure how!

Welcome to PF.
At best, you can expect about 20% quantum efficiency from an LED.
The heat will therefore be greater than 80% of total power input.
You must run the LED cool for high efficiency, and for long life.
That suggests you should design for 100% heat generation.
Vf = 2.85 V. If = 700 mA.
Maximum heat generation is 2.85 * 0.7 = 2.0 watt.
At best, that might be a 20% overestimate.

• berkeman
I write you with the challenge of calculation the heat dissipation of a LED.

The LED in question is http://www.seoulsemicon.com/en/product/spec/SZ5-M3-W0-00/22

So using the initial numbers from @Baluncore can you say what type if heat sink and enclosure you have in mind for this? Will there be multiple of those LEDs in the same fixture on the same heat sink?

Hi you two and thanks for the replies.

The cooling is already set, my task is to calculate if there will be overheating.
Roughly 400 of those LED's are used in a big floodlight.

What I need is basically the luminous efficacy of the LED (luminous flow/radiant flow). At this point I am quite sure the provided data in the datasheet is not enough to calculate that number.

My rough estimate of those LED is a luminous efficacy of 300lm/W which results in a energetic efficacy of roughly 40%. So about 60% of the total elctric power, dissipates a heat. Therefore it would be not expedient to use the assumption of 100%.

Once I figure it out I will write you share my information-

The cooling is already set, my task is to calculate if there will be overheating.
That sounds kind of backwards, no? Who "designed" the enclosure and cooling without knowing the power dissipation requirements?