Modified Medium-Voltage Arc-Flash Incident Energy Calculation Method

The IEEE Standard 1584-2002 provides a guide for conducting arc-flash hazard analyses for low- and medium-voltage systems. Users, however, need to be aware that the models in the guide were based on the measurements of arc incident energy under a specific set of test conditions combined with theoretical work. In the medium-voltage arc-flash analysis, the incident energy calculation, according to the guide, is a function of system voltage, gap between conductors, the distance from the possible arcing point to the worker, the bolted three-phase fault, and the arc clearing time from the source-side protecting device. However, the effect of fault current decaying is not considered in the current guide. According to an IEEE-recommended short-circuit calculation method, the fault contributions from generators and induction and synchronous motors would usually decay over the short-circuit period based on their short-circuit time constants and the excitation system. Therefore, the standard calculated incident energy from short-circuit point of view would be higher than the value if the effect of decaying current were considered. On the other hand, the arc clearing time according to the current guide is derived from the initial bolted three-fault current, which would result in a shorter arc clearing time. As a result, the standard calculated incident energy value from arc clearing point of view would be lower than the value if the effect of decaying current were considered. Therefore, there are two conflicting elements contained in the current guide, which may result in an inconclusive personal protective equipment requirement. This paper will present a modified incident energy calculation method based on the decaying three-phase fault current and also based on the arc clearing time derived from decaying three-phase fault current at the time of fault interruption. A sample calculation and a comparison of the results of the current method and the modified method are presented in this paper.