I looked with a spectrometer at ceiling lighting at work.
It is mostly fluorescent lamps. The one that bothers me like crazy seems to emit warmer light as perceived by the naked eye. I asked colleagues for their opinion and, although I am the only one to be affected - what a luck! -, we all agreed. The incriminated lamps are definitely softer than the bright fluorescent ones in the labs. Strange, I thought, because I would expect the warmer, the lesser blue radiation, and the easier on the eyes.
Then I measured the spectra and surprisingly found no difference. Bad (softer) and good (brighter) lamps are spectrally the same. Please do not compare intensities spectrum to spectrum, as I just held the fiber in front of each lamp without using the same distance and the same tilt (relative intensities within each spectrum are instead correct as the spectrometer is calibrated from ~ 320 to ~ 740 nm).
There are quite sharp lines at ~ 405 and ~ 365 nm. That is in the middle of violet and at the beginning of UV-A, respectively. Sounds bad, but also the good lights have such peaks. Also, using blue light-blocking goggles does not really make a big difference (light can come in from the top though).
Wondering what is the reason behind my immediate eyestrain and twitching when I get under the bad fluorescent lamp, the only thing that comes into my mind after those measurements is that it is not warmer versus colder but rather lower versus higher wattage. Can lower wattage imply an accentuation of the flicker?
Unfortunately, I cannot probe for shorter than 10 ms, so I cannot see above 100 Hz. I believe the spectrometer itself is capable of much better, and data-transfer to the pc is the limiting factor.
I glanced on frequencies of fluorescent lamps mounting an electronic ballast and they seem to be really high though, above 10 kHz, so I am skeptic I can be sensitive to that either. Also, I read such lamps start within half a second without flickering. I'd better ask someone from the facility department.
The LED lamp (black solid curve) is also "bad". There is quite some intensity at the border between violet and blue, but I am not sure that is why it is harsh on my eyes. Maybe it is flickering in this case too?
Finally, I turned the "bad" fluorescent lamp off. I can still see "bad" fluorescent light from other lamps of the same type in the office (red dashed curve below). The closer to the one turned off is about 1.5 meter away from my desk.
I have been checking if blending the bad light from the other bad lamps in the room with two desk lamps with incandescent bulbs helps (green and blue curves below), but it is very difficult to make an assessment. My eyes do not recover from strain like an on-off switch, unfortunately. And by the way, in a few days my colleague sitting under the bad lamp will return and I will have to turn it on again.
Useless to say that my eyes feels the best if I turn towards the window and get some natural light (blut curve).
On the weekend I was in a modern hotel with only LED headlamps. It was dreadful. I am really scared about what is coming next.