ian1962 wrote:
Hi - I'm new to this area and do not have an electrical background - thus pretty ignorant - apologies! I'm trying to get some help on a serious Arc Flash incident we had last week in which an electrician was burned. By the grace of god nobody was seriously injured. But I have a couple of questions relating to general practice of working on panels - hopefully someone might be able to help.
A contractor was working (under a permit) on a (2 week old) panel, when a 400V line feeding into the panel somehow came in contact with the back of the panel wall, causing an Arc Flash. The breaker in the panel was "isolated" (i.e switched off within the panel itself). But there was still a live feed to the breaker - so:
Q. can this really be considered "isolated"? (A shroud was fitted over the breaker so no physical contact was possible).
This counts as either insulated or isolated depending on the material of construction for the shroud. This is a common technique to abate an shock hazard and based on your description, shock hazard is not in the picture.
However, keep in mind that the major consensus safety standard, NFPA 70E, from 2000 to 2009, had a rather strange terminology in it. You did not consider hazards of any kind unless the component was exposed. The fact that it was not exposed means that under the strict letter of the code, there is no electrical hazard. Obviously the arc flash hazard CAN still be present. In the 2012 edition of NFPA 70E, the wording was changed to reflect the fact that shock and arc flash hazards are in fact two different, independent hazards.
Regardless, although there are tables in NFPA 70E which make specific recommendations as to the PPE requirements, in some cases these are excessive. In other cases, blind use of the tables is insufficient. An arc flash hazard analysis study is the best approach.
The 70E Technical Committee has made it clear that at least in MCC's (not necessarily panelboards) if properly designed that as long as the MCC is not changing state, and as long as certain obviously hazardous work is not going on (such as drilling holes or fishing wire blindly behind panels), the arc flash risk is minimal. However, you still have to consider the conductors in the bucket that are always energized and the probability that they will be disturbed. Your description is potentially a case demonstrating this valid concern.
What is not obvious from your description is whether or not the panelboard was powered up at the time beyond the main breaker at the top of the panel. If it was, this was just plain dangerous. NFPA 70E's requirements in this case is that this is considered energized work. When performing energized work, the first rule is not to do it unless absolutely necessary. De-energize first whenever possible. The situations where energized work would be acceptable by that code are when the voltage is under 50 volts, when the work required cannot be done any other way such as when doing voltage testing, and when working on it de-energized would create a greater hazard such as shutting off a critical safety system such as a blower that protects against release of toxic fumes. Production losses are not considered acceptable reasons. Even hospitals which used to claim the greater hazard rule have recently trended towards preference for de-energizing.
If work must proceed energized, according to NFPA 70E, a safety briefing must be held and certain specific safety items must be addressed. Anyone performing the work must sign off on an energized work permit discussing these items. Only diagnostic work (no circuit changes allowed) does not require a permit. In practice I hardly ever use the permits because the times where this activity would be acceptable are rare, even in a very large multimillion dollar mining and chemical plant. I can think of only one or two times where I have issued permits because the risk was very minimal, we went in with full PPE, and there was a greater hazard involved.
Quote:
It is being also speculated that someone must have "stood" on the cable in question which feeds into the panel from beneath and is not protected by conduit etc. and is at ankle height - and that this caused it to slip out of the breaker housing - and that this is the ONLY explanation of how the incident could have occurred in the way describer. (We haven't yet talked to the injured guy)
While this is certainly possible, there are other possibilities. One very common problem that I see all the time is improper terminations into an enclosure. For instance there is supposed to be either a plastic or grounded bushing over the end of conduit terminating into an enclosure with conduit nuts securing it (Myers hubs and some other types of terminations don't have this problem). It is very easy for the wire to chafe or otherwise rub against the sharp metal edge and cut into the insulation. This can easily even happen with the individual involved and no other person. Another possibility is that way too much insulation was stripped (beyond the stripping guideline for the lug it was terminated to) and it was bumped and made contact. Another possibility is that a few strands were not properly secured inside the lug. Either way the other obvious question is whether the lugs were properly tightened in the first place. All of these are often extremely difficult to answer after an arc flash occurs because the damage is so great that there is very little forensic evidence to go from.
Quite often I have found that panelboards have very high arc flash hazards simply because the old standard design methodology is to tie the main breaker at the top of the panel directly to the transformer secondary. At that point the only thing limiting the current is the fuse on the primary side of the transformer, that often will not even trip. This leads to extremely high ratings and to a work practice where the only safe way to work on the panel is to shut off the transformer on the primary side.
If you have one of these, the easiest solution is almost always to put a fused disconnect or circuit breaker or some other device in between the panel and the transformer.
Quote:
Q. Are there any requirements/standards to mechanically protect cabling from this kind of stress?
I very much appreciate any experience or insights you might be able to bring to bear on this
Based on your description, you are not in a U.S. jurisdiction. However as I am mostly familiar with that, I'm going to give a U.S.-centric answer. Where cables are subject to physical damage, the only wiring method allowed in the U.S. according to NFPA 70 (NEC) rules is inside rigid metallic conduit. I have a local rule that I do not accept trade sizes under 3/4". The reason for this is that even though all the codes say NOT to hang things on conduit or to use it as a ladder, people do it all the time. Below that size, putting your weight on the conduit is guaranteed to bend it. Above that size, the chance is far less. All that being said, I do accept use of armored tray cable since in practice (mining environment) it is just as reliable as rigid conduit and more reliable than mining duty portable cables in many applications. This is technically a Code violation but in that jurisdiction, NEC does not apply anyways (it is a consensus safety standard, not regulatory).
Regardless, the codes in the U.S. definitely require cables to be protected from mechanical damage in all cases. Even then, the level of damage protection required does vary. In U.S. and Canadian underground and coal mines which make extensive use of portable cables, the power cables are required to be protected by high resistance grounding, grounded shields, have the grounds woven so that they are shorter than the power cables, and the entire ground system must have a ground check monitor installed to trip out the power if it is ever damaged...the idea being that you can just about cut the cable in two with a bull dozer and the breaker will trip before major damage occurs.
Just using high resistance grounding itself drastically reduces the arc flash hazard probability by about 90% because over 90% of arcing faults are line-to-ground. Installing and maintaining such a system eliminates the most common hazard. It does not ELIMINATE the arc flash hazard. It simply makes it far less likely.
If the arc flash hazard study is done correctly, it then becomes possible to determine how to modify or adjust either work procedures or the equipment itself such as changing fuse or breaker settings, or installing specialized equipment such as "maintenance switches" or VAMP arc flash detection relays which can mitigate the potential arc flash hazard down to an acceptable level.
