NFPA 70E and IEEE 1584 use similar definitions of the Arc Flash Boundary by defining the boundary from the arc source (IEEE uses distance from live parts). Since it is often difficult to know where the live parts or source of the arc may be inside equipment, most define the arc flash boundary from the front / outside of the equipment.

This week's question is simple enough.

Do you measure the arc flash boundary from outside of the equipment?

Yes
No
It Depends

In my opinion the right way is measuring from nearest live parts, calculated from panel construction schematic diagram. If we don't have it, it is more conservative calculate from outside of the panel/switchgear until we could measure on site( in scheduled shutdown)

I don't see much difference in measuring arc flash boundary from live parts inside the equipment according to NFPA 70E and IEEE 1584, or from outside of the equipment. I believe that measuring the boundary from outside of the equipment is safer.

I see a big problem with arc flash boundary being usually calculated based on 1.2 cal/cm2 incident energy and not on actual incident energy to 2nd degree burn. Using the 1.2 cal/cm2 threshold for exposure times less than one (1) second result in undervalued arc flash boundaries. In fact, the actual incident energy to second degree burn is not a constant, it is variable function of time and it can be calculated by:

Eb = 1.2 * t^0.3,

where t is exposure time in seconds, Eb is threshold incident energy in cal/cm^2. Check please http://arcadvisor.com/faq/threshold-incident-energy-second-degree-burn for more information about the incident energy to 2nd degree burn and for free online Time and Threshold Incident Energy for Bare Skin Exposure vs Thermal Radiation Level calculator.