Do you want to measure spatial distribution of radiation from your LEDs and analyze it by spectra in each direction? If yes typical method to measure spatial distribution of light emmission from light sources is suitable. You have to replace luxmeter/irradiance meter with the spectroradiometer. Fiberoptic device would be very easy to use in this case. There are several possibilities how to construct this kind of stand. Finally, recently one of my students has investigated spectral distribution of LEDs in different directions as a part of his egineer project - in his case there were no differences, but he tested only few LEDs. I am curous your results. He used C-gamma  goniometric stand with the fiberoptic spectroradiometer.

I do not recommend camera based method. 

Thank you very much Matija Milanic, and Urszula Blaszczak for your valuable suggestion. My primary goal is now, how to utilize present LED lights to get the best spectrum that will match the D65 spectrum under day light. For this as first step, I have to measure the LEDs spectrum and lumen values. I have selected several devices including radiospectrophotometer, and so on. I will compare my data also among various light measuring devices so on.

But, as nowadays most of the office scenario, the light is on the ceiling only, in this case I only measure the horizontal illuminance? or vertical also? suppose in a standard room size, the LED lights are well distributed in the ceiling only, how should i place my spectrometer? is there any method for standard room size? should I take horizontal and vertical measurement both?

If some lights are set on the wall too, vertically, which kind of technique I will use with my spectrometer.

If I have only LED lights on wall, vertically, should I take both horizontal and vertical measurements?

( all basis on spectrum and lumens)

Dear Luke,

first of all, thank you very much for your wonderful paper, its really informative and innovative. I read through your paper and definitely you worked really hard to bring all the information come true. I am currently trying to implement your work in my thesis experiment, specially pupilary light reflex, although i am not sure my lab holds proper instruments.