iwire:
True, this is not a thread about switching power supplies, unless you count some of the new electronic ballast designs intended to eliminate flicker in fluorescent lighting. But I do not know how those units are designed nor have I evaluated them, so I simply do not know. I doubt Larry Ural's issue involves that (but I can't rule it out, either). But I have encountered harmonic issues in single phase power where loads were exclusively switching power supplies. Those issues had nothing to do with overloading the neutral.
resqcapt19:
Other than having actually experienced the problem in one of the places I have worked, there doesn't seem to be a whole lot of documentation on this. But here is one web page that shows some waveforms of loads being considered in the design of AFCI protection devices. Scroll down to number 8 to see a switching power supply. If you're going to put 100 of these in a room, are you going to use the RMS figure, or the peak figure, to calculate load?
http://www.zlan.com/waveforms.htm Whether or not harmonics is an issue in Larry Ural's case or not, I cannot say (especially since I'm not an electrician ... I just spec work and hire electricians when I need things done). I'd certainly would have suspected loose connections if he had not said everything was tight.