Here is the scenario: Incoming transformer secondary (480V) to switchboard results in 146 cal/cm^2. AF boundary of 340 inches or so. How far away would a remote control station (to trip/close the main breaker) have to be located to allow no PPE to be required while tripping/closing the breaker remotely?

Second question: The distribution board mounted next to the main gear and connected via conduit is <1.2cal/cm^2. Is this too close to the incoming to be labeled <1.2?

1. I'd guess 340 inches (or so).

2. I'd say no.

In answer to the second question there is a concept of propagation of an arc flash where it initiates in one area and then spreads to another area. This is due to two effects. The first is called arc tracking. Arcs are magnetically propelled at a speed of roughly 400-800 feet per second along busbars away from the power source until they reach a substantial "barrier" such as a bus clip, a booted opening, or ends of bus bars at which point the plasma pools and without careful examination to look for the arc tracks themselves, someone would say that the location is where the arc flash occurred. NFPA 951 (investigations) and some acaedmic papers from JHU have more to say on this. The second effect is that as air temperature increases, the dielectric properties decrease eventually to the point where an arc can occur at a new location removed from the fault. This is what is responsible for turning a single phase arc flash into a 3 phase arc flash. Thus within a panelboard you need to treat the rating (cal/cm^2) over the entire panelboard as equal to having an arc flash occur on the bus side of the main breaker within that panel. I have seen more than one case where an arc will propagate vertically within an MCC bus. But I haven't seen it occur horizontally from section to section, nor from section to section within switchgear, nor from one box to another via conduit. So no, no reason to assume that arc flash in a panel in the same room connected by conduit is different.

In clarifying the first question, obviously sometimes the math starts to look outright silly. 340 inches is 28.3 feet. There is an issue in that the exponent is 1.641 which is technically accurate over short distances but it is clearly more of an open air scenario at that distance so technically the exponent after some point should be 2.0. But since there is no research/guidance on when the exponents "level out" to 2.0, there's not much that can be done with this information other than knowledge that the arc flash boundary is extremely conservative. I'm not sure what the maximum distance is but suffice to say that IEEE 1584 test data did not include measuring incident energy at that kind of distance.

Second is that if there are obstructions can drastically reduce the distance. For instance around lime and cement kilns, they use a sheet of aluminum siding as a heat shield because it has a very low emissivity (mostly reflects radiant heat). This may not be substantial enough so you may want to consider substantial brick, masonry, or steel walls as barriers. Depending on the equipment if you have a shunt trip/close type breaker and you operate it via an SEL relay, you can get a bluetooth dongle for them so that you can operate the breaker via Bluetooth connected to your laptop or they make a handheld device that works the same way from even a few hundred feet away.

Thanks for the reply. Can you clarify above? Are you saying I can label that panel connected via conduit as <1.2cal?

I am planning on labeling the entire switchboard per the incoming bus arc flash energy.

I would say no-the distribution panel is not too close. I do not think there would be any propagation through the conduit.

For the switchboard - its in the math but if the equipment is in a "maintained" condition and buttoned up I do not think remote tripping and closing is required as per 70E Table 130.7

Couple things to consider… First, the conduit between the two pieces of equipment will act as an arc chute. If the switchboard is buttoned up while the distribution panelboard is open then this conduit could potentially amplify pressure/sound within the distribution panelboard from an event that occurs within the switchboard.

Second, if the sensors on the main breaker in the switchboard are installed on the line side of the breaker then the whole panel downstream of the incoming cable lugs should have a smaller cal/cm^2 value. This should be the maximum expected cal/cm^2 value within the buttoned up switchboard while work is being performed in distribution panelboard.

If the board only has 1 transformer feed then consider operating the medium or high voltage switch supplying the transformer. That way you can open and close as well as rack in and out the LV ACB without any PPE.

Just as I said. Looking at switchgear, even though there are openings in the back where the bus bars pass from one section to the next, I haven't seen damage propagate from one section to another. The same is also true with equipment of any kind and nearby enclosures connected via conduits. I could see it happening if the diameter of the conduit is close to the area of the side of the enclosure it is connected to but that is not generally the case.

I have not seen arc flash propagate from the medium voltage to low voltage compartments within switchgear. I don't know the dimensions of your equipment but that's probably the closest comparable to what you are describing.

I have seen arc flash propagate vertically within an MCC but not from one section to a totally different section, even though the relative size of the openings for the bus bars where the splice plates are located is a lot larger. So this should give you some idea of dimensionally where an enclosure is and is not a separate enclosure.