IN the 2009 version of 70E 130.3 is the "call for the Arc Flash Hazard Analysis. The following language was also included:
[font=Calibri]Exception No. 1: An arc flash hazard analysis shall not be[/font]
[font=Calibri]required where all of the following conditions exist:[/font]
[font=Calibri](1) The circuit is rated 240 volts or less.[/font]
[font=Calibri](2) The circuit is supplied by one transformer.[/font]
[font=Calibri](3) The transformer supplying the circuit is rated less than[/font]
[font=Calibri]125 kVA.[/font]

Since this exception text is no longer in the 2012 version of 70E, should one conclude that even smaller systems that meet this criteria now must be considered for an arc flash study. I did not see any other areas of 70E text that "re-stated" this language.

This exception ONLY existed in one of the 8 arc flash calculation methods given in the annex. It should not have been in there because use of the exception depends on the method used. This was crossing the line between specifying how to do an arc flash hazard evaluation and work rules.

The exception still exists in IEEE 1584 today. It is going to change in the near future. The exact language of a "lower cutoff" is not clear yet. You can read about some of the preliminary results by looking on Mersen's web site under their technical articles. They have two or three published this year alone that make a strong case for why you can't categorically ignore cases under 240 V/125 kVA.

Here is a bit more information related to this subject that I posted recently.

http://www.arcflashforum.com/threads/1830/page-2#post-12121

It appears that in the new version of IEEE 1584, the level of 125 kVA may be reduced downward to 45 kVA. This is because experiments have shown that an Arc Flash is sustainable for 75 kVA transformers at 208 VAC.

Thank you for the information. So you are saying it is possible to sustain an arc at 208 VAC when fed from a 75 kVA transformer. But what is likely, common, practical, realistic, and applied to the working environment? I am trying to get a feel for what the industry is doing. Are we now excluding studies on systems fed by the lower 75 kVA xfmr value? Or are most people still using the 125 kVA rule?

Old thread but some work at a PG&E lab said no. It was later found that the standard IEEE 1584 test that uses a 12 gauge wire as a fuse does not initiate an arc flash but a smaller wire will especially with electrodes very close together (but not too close) when combined with placing barriers to concentrate heat/plasma. So we know it is possible. Whether or not it would happen in real world cases is not known at this point, and the threshold has not been quantified, nor proven by an actual case. So far no temporary update or revision on IEEE 1584 has been published. So there is no revised consensus standard. Keep abreast of research but don't take it as gospel except for research purposes.

Funny how the NEC Committee can argue on "show me the fatality" grounds but not 70E.