Originally Posted by Nyal Mellor
...where are you getting the 600V figure from?
Martzloff and others. It has been so long now, I cannot remember all the sources. Intel had a requirement for power supplies to be used with its board designs to be able to withstand a 800V or so surge IIRC. Business computer manufacturers built to a similar standard.
I could also give you some personal annecdotes. For instance, before I installed a whole-house device, I was replacing incandescent lamps and dimmers at an excessive rate. After installing the whole-house device, the problem disappeared. It takes a 800V or greater surge to kill an incandescent lamp, 1200V or so to kill the triac in a dimmer. Yet, I have never had a failure of AV gear before or after installing the whole-house device. IOW, it seems that my AV gear was inherently immune to surges of at least 800V or more.
Interestingly, the fact that we are not replacing incandescent lamps all the time means that surges of 800V or more are rare events, this according to Martzloff.
330V peak (2x normal) is the IEEE Emerald Book recommendation in terms of how much to reduce the surges to.
I don't know that it is recommended to reduce surges to 330V, but that is the minimum rating that can be assigned to MOV-based surge protective devices under UL1449. And IIRC part of the definition of a surge is that it is at least twice the nominal voltage.
...seems to be commonplace across a lot of industries including IT for example.
I wouldn't put a whole lot of stock in what IT does when it comes to surge protection, even in the biggest companies. Their expertise is computer hardware, database systems, operating systems, networks, etc., not power quality. Ever see a power quality engineer in an IT department?