Modeling DC Arc Physics and Applications for DC Arc Flash Risk Assessment

Although the physics of low-current, DC arcs has been studied for over 80 years, a focus on high-current, DC arcs began about 1970, primarily for understanding low-voltage, stable arcs for welding, arc lamp, and arc furnace applications. These arcs are dominated by energy conversion to radiated energy. A more recent interest in understanding longer, higher-current, and higher-voltage DC arcs began only about 20 years ago with the concern of DC arc flash hazards. For these arcs the dominant energy conversion results in an expanding plasma, the arc flash. AC arc flash hazards were recognized in 1982 and introduced into OSHA and NFPA 70E soon thereafter. DC arc flash hazards were being considered by 2007 and introduced into the 2012 NFPA 70E. Two methods were introduced in NFPA 70E to estimate DC incident energy, with little data available. In general, these methods overestimate the hazard when compared to recent laboratory data. This paper will provide a high-level overview of (a) the current understanding of high-current, DC arc physics, (b) models presently used in performing DC incident energy analyses, and (c) recent laboratory studies to measure DC arc flash parameters for several voltages. This latest information will then be used to evaluate the accuracy of some existing DC arc hazard assessments, and to propose improved approaches to using nonlinear models, based on the nonlinear nature of the DC source, for more accurate analysis. This paper will consider the unique modeling needs of large battery systems (lead acid, lithium ion, flow, etc.), capacitor systems, and solar voltaic systems.