arcee93 wrote:

i have a 600A main breaker panelboard on secondary of 480D-240D 225 kva transformer. Short circuit current is 8955 A. I'm getting 701" Arc flash boundary and 488.8 cal/cm2. Trip time in report shows 34.42 sec. Does this seem right? Am i modeling something wrong? or maybe the relatively low fault current would allow an arc to continue for a time before the overcurrent trips?

OK, there are a few problems here not the least of which is the calculation itself and whether or not the calculation is even valid in the first place.

In a scenario such as a vault where "escape" is not readily possible using IEEE 1584 calculations, this might be close to correct except that in that same scenario the fact that the arc flash would be "contained" indicates that even higher values are possible. In this scenario, arc flash blankets with actual testing is the best approach.

In normal conditions where "escape" is possible, IEEE 1584-2002 gives a cutoff value of 2 seconds. There is a lot of verbiage around why this value makes sense but it has become an industry standard. Using this value, the incident energy drops to 28 cal/cm2. Now we're at least in "40 cal suit" range, but we're not done yet.

There is also the troubling aspect that somewhere below around 250-300 V, arcs tend to self-extinguish. IEEE 1584-2002 calculations are only valid for stable arcing faults which don't self-extinguish. IEEE 1584 addresses this for transformers rated 125 kVA and less at 208 V and less but you are outside of that scenario. IEEE C2 (NESC) 2012 and later gives values based on actual equipment testing of 4 cal/cm2 for the scenario you've described. Since we don't yet have a formula for self-extinguishing or "weak" arcs the table approach in IEEE C2 is really all that is available to address low voltage arcing conditions above the IEEE 1584 "125 kVA rule". I believe the actual testing in that scenario used a 20 kA available fault current and the highest measured incident energy was 3.2 cal/cm2, well below that predicted by IEEE 1584.

Since actual equipment testing trumps calculations, I'd throw out the calculation especially in this case where the calculation is an unrealistically high value with or without the 2 second rule.