An arc flash study has been performed and has identified some gear and switches as Dangerous (>40cal/cm). When a cabinet is closed, is the hazard level reduced?

Can a person safety operate the switch?

What PPE is recommended to perform such a task?

The probability of an unintended interaction (dropping a tool, bumping a panel phase-to-phase or phase-to ground) is reduced with the doors closed/panels on, but the arc flash energy will not be reduced in an incident occurs, and may, in fact, be worse, since the doors may become part of the flash plasma or may blow off and increase injury

[/quote] Can a person safety operate the switch? [/quote]
Stating the obvious, if the circuit is de-energized upstream, then the switch can certainly be operated safely; otherwise, from within the arc flash boundary, no, it cannot although many will disagree. Switching a properly maintained switch has a very low probability of catastrophically arcing, but high energy switching (NFPA 70E goes as low as 600V) requires PPE. I look forward to what others think, because there will be shops that do not require PPE for switching with doors closed.

[/quote]What PPE is recommended to perform such a task?[/quote]
There are suits up to 100 cal, but the blast becomes the issue and might not be survivable.

I think that might depend on the switchgear. I have seen some newer arc resistant switchgear that is designed with blowout panels to channel the blast away from the operator. Some (I have specifically seen Arc Armor Enclosures) are advertized that the control compartment is designed for minimal PPE to operate. At the time I was looking into conpartmentalized enclosures for an application where the feed from the grid was 50 - 75 cal/cm^2



I am not sure why someone would design a suit where the only assurance would be that corpse would be presentable.

If you have a 100 cal/cm^2 incident, aside from the fact that the average human body can only withstand 50psi instantaneous pressure differential, you can be assured that the blast pressure will likely far exceed that and knock you back perhaps 50-100 feet, possibly into a concrete wall.

Did the study make any recommendations to reduce the AFH? Is there a breaker upstream of the switch with adjustable instantaneous settings? Sometimes lowering the instantaneous setting can lower the AFH.

An arc blast is only dependent on the available fault current, and not on the time it takes to clear that fault. Therefore you could have a high available incident energy due to a long time to clear the fault (say 2 seconds), but there may not be a large blast effect.

You need to perform a hazard analysis, which takes into account the severity of the flash, the probability that it will occur, the condition of the equipment, and other factors. See Annex F of 70E for more info.

There was a flash a couple of weeks ago outside Washington, DC. An electrician for a service company was called in to look at a 4000A fused disconnect switch that was having problems (don't know the exact details). When he opened the switch, with the doors closed, there was a flash, and ended up with a couple week stay in the hospital. So it definitely is a possibility that there be an issue when operating a switch with the doors closed, as it happened in my neighborhood.

Here is a short article I wrote a while back about the mis interpretation problem with the 40 cal/cm2 value. We don't have any good answers yet but as a few mentioned above, blast is more of a function of energy per time (short circuit current) rather than total energy.
[url='http://brainfiller.com/library-articles/arc-blast-and-40-calories-centimeter-squared.31/'][Article][/url]