I found a very interesting paper, that developed an empirical formula to calculate the MPE (s) (maximum permissible exposure limit) adapted to conditions that are common for desk lamps. You supposedly only need to know the luminous flux (lm), distance (a question will follow) and color temperature (K) to enter them into formula 8 or you can estimate using a given table of values.
- The spectral wavelength composition obviously plays a role here, but only "three light sources, i.e. OLED, LED and CFL" are mentioned here without further specifications, am I missing something? After all, reference is made here to the HEV spectrum, for example, which is of course different depending on the LED and phosphor coating...
- What I also find strange is that absolutely no account is taken of the diffusion properties of the surface (like the book or other materials in figure 2, I can hardly imagine that this is so insignificant... but ok, it's an ideal, 100% reflection assumption)
- It's strange that Figure 2 refers to "viewing distance", that would mean only the distance between eyes and surface, whats about the distance from the lamp to the surface? My lamp, for example, is just as far away from the table surface as my eyes, which leads to a doubling of the calculated distance and a corresponding significant increase in the MPE value. I find this very unclear.
- Hard to work with photometric quantities like lumen and lux here, it would make more sense (and be easier to understand) to stick to photon flux and irradiance (mW/cm^2), then you could also take out "color temperature" (which I think is just a naturalized term for the color appearance of a black body at a certain absolute temperature, but does not have to follow its spectral composition at x K, the main thing is that it appears in the same color as the black body would to humans [that would also be bad if my 5000 K desk lamp emits UV rays at peak. ..]).
Again to point 1, you can already notice the difference between the fact that significantly higher lux values are achieved outside (100 to 1000 times more) than inside, according to these MPE values that result from this, we should all have gone blind long ago. In general, most LEDs naturally have the property of having a peak in the blue range, which is not the case with the solar spectrum. That said, I find it strange that nothing is written about the spectral composition of the lamps that were tested, after all "color temperature" does not equal "spectral composition" (except for an ideal black body).
For example, that would make the formula totally pointless if the desk lamp [like many of these days] has an extra absorption filter built in for the HEV range, but they state, it's supposed to be "for any given desk lamp at any viewing distance."