All -- We have many areas where we need to interact with breakers of varied ratings (60k AFC+/-) and we do not have sufficient arc flash information to know we are donning the proper PPE. In these cases we are attaching/tying/taping string or ropes to the breaker handles and then running the rope outside the room to remotely operate the switch. As I understand things; provided the assembly between you and the equipment is "substantial" this practice is ok. Can anyone help me with defining "substantial"? I thought something about this was in the code but I can't find it. Thanks in advance.

We also use ropes and pulleys for the same reason. When first creating the mechanical set up we contacted the circuit breaker manufacturer. The manufacturer didn't have a heartburn with it as long as we didn't in any way modify the breaker, handles, case, etc. You should contact the manufacturer for further clarification. Depending on the type of breaker, if it has an automatic recharging spring, you may need to consider the potential for incidentally initiating the closure if a push button is bumped while installing the apparatus.

First off, AIC has nothing to do with arc flash directly. Arc flash is a very different effect from a bolted fault. During a bolted fault, the major engineering issue is that the magnetic forces on the bus bars tend to rip things apart. With an arc flash, the major issue is that the air heats up to the point where it becomes a conductor and releases large amounts of heat and/or plasma. These are two completely different effects.

Considering all of the available arc flash hazard analysis methods if you have ZERO information, you would best be served by using the tables in 70E. They are by no means the best and there have been cases where personnel have been injured while following the recommendations in the tables but until you get the arc flash hazard study done, this is the best you have to go on.

Now to answer your question more directly...this is another one of those grey areas where you won't really find anything. So far the research has been mostly on studying stable arcs in typical industrial equipment. Some data has been collected on the arc blasts caused by an arcing fault but there is no available guidance or standard on the subject yet due to lack of comprehensive research.

For the thermal (and plasma) energy released by the arc flash itself, pretty much anything that has a fire rating will work. The arc flash is typically very short lived (we use a rule of thumb of no more than 2 seconds for most cases) so the amount of heat that can actually pass through a structure will not be very much. The bigger concern is whether the structure is flammable or not, so almost any steel/concrete/gypsum structure passes this test. Steel can burn through at close proximity to an arc flash (such as the enclosures of the electrical equipment) but this is at very close range. Similarly concrete can and does "burn" (part of the reason for the collapse of the World Trade Center buildings) but only at very elevated temperatures for a prolonged period. Gypsum works until the water evaporates.

There is as I said, very little information on arc blast. There is a theoretical equation by Lee which claims to predict the arc blast pressure but I have seen far too many real world examples where the Lee equation is clearly invalid.

At this time the best recommendation I can give you as to construction advice beyond this would be API RP-752 and RP-753. These standards were created in response to a well known major oil refinery disaster in Texas City. A flaring tower exploded and destroyed several construction trailers that were set up right next to it. The explosion killed 15 people and injured about 180 more that were in or around the tailers. The magnitude is nowhere near where even the largest arc flash could come even close, but this is a very generic standard that specifically addresses issues of building construction and design to withstand damage caused by concussive pressure waves.