Hello everyone, I have been asked a question that I am not sure of the answer. In a facility I have been in recently there are 13200v to 480v transformers present. In the transformer enclosures there are cooling louvers. What is the general opinion on cooling louvers as concerned to arc flash? Is having louvers present the equivalent of having exposed live parts even though the covers are in place? Thanks for any opinions on this matter!!!!

It's task specific. There are no exposed live parts unless you have someone sticking bare conductive materials into the vents. It is the same thing as blind reaching or sticking metal fish tape into a hole in something without knowing where it can go. Visibility of the energized part is not a prequisite to being exposed. Recessed screw heads for instance on a breaker are certainly visible but by design usually not considered exposed, unless you are sticking a screw driver down inside the holes to tighten the screw while it is live!

Don't confuse the arc flash hazard with the shock hazard. An arc flash hazard occurs because the activity that is occurring is likely to cause an arc flash. Once it occurs, the various methods to calculate the arc flash hazard pretty much ignore the equipment doors and walls. So it really doesn't matter whether there are louvers there or not because the flash hazard boundary is the same. There are some "common sense" views on this though. Equipment enclosures generally don't protect against anything unless it is arc resistant but a 2-4 hour fire wall certainly offers pretty good protection such as a concrete block wall and should be considered a substantial barrier to an arc flash.

On a similar note, as someone might be asking this, is about putting a Lexan barrier about 3 feet from cooling vents on transformer and switchgear enclosures. The thought being that if an arc flash were to occur when someone was walking by, they would be protected.

Thoughts/comments/opinions?

It's not the first time I've seen this recommended and/or done. I disagree that the concept or the implementation is valid.

1. I have looked through hundreds of OSHA accident investigations using the key word "electrical" which captures shocks, arc flashes, and a few cases that have nothing to do with an "electrical" hazard. OSHA did not report any cases of spontaneous arcing faults which caused someone to be injured or die. In fact less than 10% of the cases were even those where the workers involved did not do something clearly dangerous. I'm not ruling out "spontaneous" arcing faults. But I'm not ruling out meteor showers and global thermal nuclear war either. All of these events are possible, but the likelihood is really remote.

2. There have been several times where the 70E Committee has stated over and over again that "just walking by" is not as it is currently defined, an "arc flash hazard" because "just walking by" is not interacting with the equipment in such a way that it creates an arc. The Committee recognizes the same thing. There have been numerous attempts to quantify this but so far the wording fails to capture the concept. Things will improve in the 2015 edition in this regard though.

3. So leaving aside that this is a non-issue, let's consider whether or not the Lexan shield will actually have any benefit whatsoever.

First, we have the arc blast to contend with. You can't realistically contain any kind of "blast" except very small ones in small areas. As the size goes up, the amount of framing goes up substantially. That's the reason why costs on large explosion proof junction boxes rapidly reach a point where they become uneconomical. So with a large box surrounding a transformer constructing it strong enough would be a big challenge, certainly not one where you can just screw together some Lexan panels.

Second let's consider what we are trying to do. Arc flash as it is currently modelled is a case of extreme radiant heat. It's mostly infrared so the lexan solution is meaningless. You'd need something tinted, or solid. With transmission rates in Lexan being over 90%, it won't stop what we refer to as an "arc flash", even if it holds up.

And finally, hopefully the Lexan will pass all the radiant heat along because it melts and may become a bigger hazard than the one it is meant to contain. Even if it passes most of the radiant heat on (what it was meant to do), any amount of dust or dirt would quickly make it the same problem.

Perhaps a more realistic "shield" would be using an arc flash blanket. Again the matter of creating a substantial barrier becomes an issue. But even then, the manufacturer(s) all recommend wearing FR PEE anyways since the blanket reduces but does not totally eliminate the arc flash hazard.

4. I'm an engineer though. I am not an HR person and sometimes I recognize that the most ridiculous and silly notions win out. So if that's the case, why not just move the transformer or move the people? The technological solution is more complicated. Put a breaker on the primary side of the transformer that has a shunt trip input. Put bushing CT's on both the primary and secondary lugs of the transformer. Buy a big, expensive relay that can do both 50/51 and 87 relaying and trip the breaker. This is pretty common for transformers over 10 MVA but there's no reason it can't be done on smaller transformers these days. With this arrangement, trip times will be equal to between 1/4 and 1 cycle plus whatever the opening time of the breaker is. Getting down to about 0.065 seconds tripping time is not unrealistic even for a large medium voltage transformer, and the arc flash hazard at that point will probably be under 1.2 cal/cm^2, enough that the arc flash hazard boundary is inside the transformer enclosure. It could be a straight 50/51 relay if the only concern is external faults (which it usually is) but if internal faults need to be detected too, the 87 relay is pretty much the only way to go.