Hi everyone.

Some of the equipment we need to complete arc flash studies for have vertical insulating barriers installed between each phase (for example - 12.47kV-600/347V live front padmounted transformers [photos attached] and 15kV outdoor metal enclosed switchgear). My question is, in these cases, should we be considering the direct electrode to electrode gap as if the glastic barriers were not in place or should we be considering the electrode gap to be the shortest distance between electrode taking the path around the glastic barriers?

I have tried doing a good amount of research but cannot find anything discussing one way or the other. I've found many articles which mention insulating barriers holding up quite well, being charred but not burned through during arc flash incidents, but have also found a source stating the resin in glastic barriers melts as low as 232degC (450degF). The one concern I see with considering the longer path is that it will almost certainly exceed the IEEE 1584-18 maximum bus gaps, resulting in needing to use the less favorable/ accurate Ralph-Lee method (Milsoft, online calculators, and the IEEE spreadsheet will not perform the analysis if the parameters are outside of the IEEE 1584 bounds).

If anyone could provide their thoughts or experiences, it would be greatly appreciated.

Thanks!

Serge

As you found, there is still limited information for every arc flash situation. IEEE 1584 assumes there are no barriers between the bus.

Many would assume "worst case" with no barrier with the reasoning, what if it was removed, damaged, melted - basically some justification for calculating a straight line between the buses. Could that happen? Who knows but when legal minds get involved....

Others may have different views, but this is a pretty common one.

How does it affect the incident energy? Remember in the end it is all about making sure the worker is protected from the arc flash should there be energized work (which it to be avoided unless justified based on NFPA 70E)
Do the different assumptions result in higher arc ratings for PPE?