Edward Murphy is famous for his law that states: “Anything that can go wrong will go wrong.” When performing an arc flash study, Murphy’s Law becomes very important if assumptions are made about this rare but potentially deadly event.
An endless stream of videos show just how devastating an arc flash can be. Scenes include horrific “fireballs” shooting out of equipment, doors being blown off and, of course, the tragic stories about the arc flash victims.
Two questions require assumptions (and can still spark debate):
Assumption 1—doors and enclosures
When an arc flash occurs inside electrical equipment, there is an almost instantaneous pressure buildup from the rapid expansion of vaporizing conductor material and superheated air from the arc. Depending on the magnitude of the pressure compared to the construction (strength) of the enclosure and doors, a failure could occur. However, even if the enclosure remains intact, the energy will attempt to escape through any opening—such as vents, joints or cracks—sometimes violently. Yet, not every arc flash results in doors blowing open, an enclosure failure or energy “escaping.”
When performing an arc flash study, how can you determine whether an enclosure will contain the arc flash? That is the problem. There is simply not enough data available to make this judgment with any degree of certainty. Unless the equipment is designed as arc-resistant, there are no assurances that the doors will remain closed or that the enclosure will not fail. Arc-resistant equipment, when properly used, is designed to contain the energy of an arc flash and vent it away so the worker is protected.
During some of my more recent arc flash testing, we experienced both sides of this issue. The tests included staging a three-phase arc flash on three different 480-volt (V) switchboards. The available short-circuit current was 50,000 amperes, which is typical of what may be experienced at the 480V secondary of a 2,500 kilovolt-ampere transformer. Two of the tests resulted in nothing more than the expected fireworks with significant internal damage to the equipment (this stuff never gets old). In each case, the enclosure remained generally intact, and the doors did not blow open. There was some external scorching (as energy escaped through ventilation openings and joints) and bulging of some of the enclosure steel. The third test was much more impressive because Murphy was helping out. Using the same test setup and similar switchboard construction, part of the cover blew open as shown in the photo at left.
Assumption 2—arc propagation to the line side of a main device
Another common assumption involves whether the arc flash can propagate to the line side of a main protective device as in the figure above. Industry consensus at the moment is that, depending on the type of equipment, it is assumed the arc flash could propagate to the line side of the main. Would the main trip in this case? Probably. Would it interrupt the arc flash? Not if the arc is on the line side of the device. The assumption is that the arc flash duration is defined by the time it takes the next device upstream to operate, not the main of the equipment itself.
The reason the “type of equipment” is important with this assumption is that most agree, if an arc flash occurs in a feeder cubicle of equipment such as switchgear, it would not likely propagate to the main cubicle. The same cannot be said for equipment like panels and switchboards.
During an arc flash, the arc tends to run away from the source, not toward it. This could lead to the conclusion that the arc could not possibly propagate ahead of the main, since the arc would need to run backwards. However, the conducting plasma cloud that is also formed can produce interesting results.
In another series of arc flash tests conducted earlier this year, three separate motor control center sections were used. Although the arc ran away from the source toward the bottom of the equipment, the plasma cloud caused the arc to propagate and restrike near the top of the bus.
How can you determine whether the doors and equipment will remain intact or whether the arc flash will propagate to the line side of the main? At present, you can’t. Keeping Mr. Murphy in mind, consider the worst-case scenarios when making assumptions.