source intensity vs spot intensity

[stein09163]

hi..........i wish light emitting diode data sheets would include both measurements: mcd and mw/cm2. it would make it much easier for identifying which visable led you need for your particular purpose.. i'm sure many of the led engineers out there would agree. it would save them a lot of hassle. i'm hoping i can get some help.with a answer to: if i want to use a 5mm 660nm extremely high spot intensity light emitting diode with abeam angle of 20%; i have been looking and the highest intensity i can find is 2700mcd (which is a measurement of source intensity). i'm looking for one that will give me close to 1600 mw/cm2 (measurement of spot intensity). i know it is out there but all the data sheets i have seen use mcd. just maybe if measured properly, the 2700mcd 660nm beam angle 30 degrees and spot 1" from source in 60 secs, visable led is rated close to what i'm looking for. i believe i've read that there is a equation for solving this problem but i can't find it. if there is a led engineer out there with some compassion, i would appreciate the help. rick

 

[DIWdiver]

You probably know most of this background, but let's get it on the table for readers who might not.

The candela is a measure of luminous intensity, not radiometric intensity. That means that the candela is weighted to the sensitivity of the human eye. The actual definition is a bit wordy, but it boils down to one lumen per steradian. The corresponding measure of radiometric intensity would be a watt/steradian (or perhaps mW/sr). Because of the weighting applied to the candela, to convert from one to the other requires both knowing the spectral density of the LED emission, and the weighting curve applied.

The weighting curve looks vaguely Gaussian, but it drops to zero at the infrared and ultraviolet regions and beyond. The peak is a particular wavelength of green, at 683 lm/W. White light ranges depending on the spectrum, but it's generally between 250 and 300 lm/W. In my calculations I use 273 lm/W, and assume that's probably close enough.

For various monochromatic wavelengths, you'd have to find the curve and look up the value at that wavelength. There's a graph on the Wikipedia page for 'candela'. It shows that at 660 nm, you get about 55 lm/W. In deep red, deep blue, and violet they start switching from publishing luminous intensity to radiometric intensity, because the luminous intensity is dropping toward zero. For infrared and ultraviolet, they have to use radiometric, because the luminous is zero.

Now to calculate irradiance, which is the value you want, you have to know the distance from the source to the target, because this measure drops off as the square of the distance, just as illuminance does.

So, to convert from mcd to mW/cm^2, first divide by lm/W to get mW/sr. Then divide by the number of cm^2 in a steradian at the distance you are interested in, and you have irradiance in mW/cm^2. I'll leave the last part of the calculation to you.

 

[frascati]

Posting this to another sub.