Comparative Analysis of Experimental DC Arc Flash Results to Industry Estimation Methods

Overview

The research question is how feasible and how severe DC arc flash really is, and how well industry estimation methods match it. At about 130 VDC, the voltage class of battery control and substation DC systems, the authors ran scouting tests varying bolted fault current from 1.9 to 27.5 kA and bus gap from 12.7 to 63.5 mm, and document the setup, procedure and system parameters.

Measured incident energies and arc sustainability were compared with the NFPA 70E and Doan maximum-power models, the Stokes-Oppenlander and Paukert arc-resistance models, and results from ETAP, a widely used arc flash study software. The paper also compares measured arcing current with the assumption made in NFPA 70E.

Results: only one test gave incident energy above 1.2 cal/cm²; arcs sustained across the whole fault-current range but not beyond a 25.4 mm (1 in) gap, even at 27.5 kA. Incident energy was inversely related to gap, and in one case doubling the gap from 12.7 to 25.4 mm halved it. Maximum-power models differed greatly from measurements; arc-resistance models were closer and usually conservative, although at 1.9 kA measured energy sometimes exceeded them.

Limits: the tests are scouting tests at a single voltage class (about 130 VDC) with a limited number of runs (nine in the arcing-current comparison), so they cannot be transferred directly to higher voltages such as 400-800 VDC rack busbars. They do add public DC arc flash test data and remind designers that low arcing current can lengthen clearing time and make the hazard worse.

Key points

1. Tests covered about 130 VDC, bolted fault currents of 1.9-27.5 kA and bus gaps of 12.7-63.5 mm.
2. Only one test gave incident energy above 1.2 cal/cm²; arcs sustained across the fault-current range, but no arc sustained beyond a 25.4 mm (1 in) gap, even at 27.5 kA.
3. Incident energy was inversely related to gap: in one case doubling the gap from 12.7 to 25.4 mm halved the energy.
4. NFPA 70E assumes arcing current is half the bolted fault current; the measured average was 11% to 48%, and 7 of 9 tests were a quarter or less. Lower arcing current can lengthen clearing time.
5. Maximum-power models differed greatly from measurements; arc-resistance models were closer and usually conservative, but at 1.9 kA measured energy sometimes exceeded them.
6. The closeness of arc-resistance models was attributed to their consideration of bus gap and the large effect gap had on incident energy.

Using it in design

• Keep bus gap as an input in arc flash calculations for DC panels and battery busbars; do not rely on the maximum-power method alone.
• Instead of assuming arcing current is half the bolted current, check in protection coordination whether low arcing current lengthens clearing time.
• For low-fault-current DC branches (around 1.9 kA), support model results with measurements or manufacturer data, since models may not be conservative there.

Who should read it

Electrical safety engineers labelling arc flash and selecting PPE on batteries, DC buses and DC distribution boards.

Document contents

1. I. Introduction
2. II. Experimental Set-up and Test Procedure (A. Test setup, B. Test procedure)
3. III. Test Results (A. Arc flash sustainability, B. Gap distance impact on incident energy levels, C. Peak fault current impacts on sustained arcs)
4. IV. Incident Energy Results Based on Theoretical Methods
5. V. Comparison of Incident Energy Models with Test Results
6. VI. Comparison of Arc Current Magnitude of Models
7. VII. Conclusion

Related standards

• NFPA 70E
• IEEE 1584-2018

Read the document

Publication details

Authors

Corey S. Weimann; Robert J. Kerestes; Brandon M. Grainger

Publisher

IEEE Open Journal of Industry Applications (Institute of Electrical and Electronics Engineers (IEEE))

Year

2020

Volume

1

Pages

181-193

Document language

English

Source type

Scientific

Access

Open licence — downloadable

DOI

10.1109/OJIA.2020.3031768

Licence

CC BY 4.0 — share with attribution

Licence evidence

ieeexplore.ieee.org

Citation