We have 480V lighting in our plant. And when we separate the neutral to work on it we have 277V on the neutral. I know the arc flash there has to be very low, but how do I determine what it is?

At present, there is no universal industry accepted method for determining the arc flash incident energy of single phase circuits.

Technically there is no published method. Practically, single phase faults are not drastically different from three phase in the studies that looked at this. There are some papers on Mersen's site on this. If you run a study with a 277 volt secondary, three phase transformer, the results will be higher than in the real world but not a huge amount. You will overestimate the danger but that is better than the alternative. The major issue is going to be that it is difficult to detect and trip single phase faults unless you use single phase breakers/fuses. All sensitive Earth fault designs require balanced 3 phase loads and will see your lighting loads as faults.

You really have no chance of a sustained arc flash. You may get a short circuit and subsequent flash. That is short lived and cannot be defined according to any existing arc flash calculations.

Furthermore, If you review NEC Article 210, HID lighting is limited to 50A circuits. It is highly unlikely that you will ever see greater than 1.2 cal/sq cm on that size OCP. Lastly, arcs only maintain themselves on three phase circuits because you never have a voltage zero, therefore, there is always voltage available to maintain the arc, not so on single phase circuits, they should self extinguish.

I would bet if you took a census, there has never been anyone burned at 18" from the source on a single phase circuit that is 600V or less. Virtually all the reportable accidents where people are burned are three phase and large circuit sizes.

That was the thinking before. Recent testing brought to the IEEE 1584 Committee on the issue of <240 VAC has shown otherwise. It is rare but possible.



Uhh, no. Unless the OCPD is current limiting, you can get almost an arbitrarly high current and thus incident energy can be arbitrarily high. Granted, it is highly unlikely from a practical point of view that anyone is going to have a 10 MVA single phase 120/240 V transformer, but you can't just dismiss it based on the size of the OCPD alone. If however you look at it from a very basic point of view such as using the short circuit current only using a 2 second cutoff for time using say Lee's method, this gives you an extremely conservative value that can nevertheless make the calculation easy to solve without use of high end power system analysis software, and gives you a quick way to knock off lighting panels as being a minor concern.



I highly disagree. The concept is sound but the wording here is a bit off. For one thing, as that voltage gets above 150 V (which often shows up as the arc voltage in experiments), the likelihood of a sustained arc goes up dramatically. The likelihood which you were previously arguing for a 600 V panel is pretty darned likely if it's 480 V or 600 V, SINGLE phase. We're no longer in the realm of 120, 120, or even 277 V panels when you make the upper cutoff at 600 V.

Second, there's a Youtube video showing a burn on a lighting panel. I've even been burned by one in a residential setting. However, the total damage in my case was that I got a small 1st degree burn. This is considered an acceptable risk in the way that 70E is currently structured (focus is on preventing fatalities). On the other hand if your focus is on fatalities due to burn injuries from 120/240 V lighting panels, I'm not aware of any.

Third, a census has nothing to do with injury data. It is concerned with the number of living people, not the dead and not how they died.