Those sources are exaggerating. All LED begin as blue LED and therefore
emit a small amount of UV. If the phosphor is compromised, the amount can increase. The fact that turbodog
specified purple guarantees there is UV being emitted, and we know it isn't actually purple because purple is a color mixture. Violet is the spectral color. And unless turbodog's floodlight is a freakin' violet laser, it is beyond all doubt emitting some UV.
Look, I realize I'm not popular, but it's crazy that I'm so unpopular that if I believe the truth, some will insist on believing utter crap. LED emit a small amount of UV, shown by a
Google image search for "LED color spectrum" where every single germane image shows UV emission, and everyone is just going to have to deal with that. Because in ten million years I would never care whether anyone agreed with me or not. The crowd is untruth.
idk if you're blind or what, but the spectral graph used there clearly shows shows emission down to at least 325nm, well below visible light in UV-A, and that lamp may even have more. The poster is focusing on the amount of blue and green, which in comparison, the output of UV is negligible, but it is not nothing. And the lamp's spectral output will change over time. How is anyone's guess, but turbodog's anecdote strongly suggests that with the spectral shift, the amount of UV emitted can increase.