I see that 208V is to be considered in an arc flash if it is fed from a transformer greater than 125KVA.

Are there any cases where single phase should be considered? ie, at a certain voltage level or fed from a certain size transformer, etc?

Thanks!

I think that the issue is that there are no valid models for a single phase arc flash incident. The current models are based on 3 phase only and most likely the energy in a single phase circuit is low since the circuits are usually low amperage and therefore small breakers with low instantaneous trips, therefore low energy.

The guideline says that you don't have to analyse circuits 240V and below fed from 125KVA and smaller transformers. This statement does NOT exempt single phase loads. So single phase loads have to be considered for analysis whenever they are greater than 240V and fed from larger than 125KVA Xfrms. Being exempt from analysis does not make them exempt from requiring FR clothing. I would recommend that until a good single phase model comes out, to use the three phase equations. The IE numbers would be higher than required but consequently more conservative. The breaker or fuse clearing time shouldn't be a problem as I would think the arcing current for single phase would be the same as three phase.

IEEE 1584 states that "Single phase ac systems and dc systems are not included in this guide." Additionally, the model for the energy calculations in IEEE 1584, state that the voltage is 3 phase.

In Annex D of NFPA 70E also states that the system limits for calculating incident energy is 0.208 kV to 15 kV, three-phase.

Currently there is no industry approved, valid method for calculating arc flash hazards for anything other than 3 phase systems.

The closest I can see for anything that could be used for single phase systems is in Table 130.7(C)(9) NFPA 70E, in the first section that is Panelboards or Other Equipment Rated 240V and Below. Most single phase systems will meet Note 1. The highest category is #1 for PPE.

It is probably best that a company has a minimum standard of FR clothing for any of their electrical workers, like Category #1 or #2. This would protect them on single phase systems.

Whether of not that portion of the system needs to be considered, it is the supply transformer connection that determines the relationship between the 3 phase and phase to ground fault current. Wye-Wye conection would generally lead to less single phase current, but if it is Delta-Wye, phase to ground is most likely higher because of the lower zero sequence vaules.

Alan

Thanks everyone for the replies!...

Hi

IEEE 1584 5.1: Ranges of models:

(About Lee equations)
«Where single-phase systems are encountered, this model will provide conservative results.»

I agree with Haze10.

Single-phase model

Isn't the ArcPro model a single-phase, open air model?

Yes but the ArcPro model is designed for overhead utility configurations of > 1000 V as I understand it since it is the basis for the Table 410-1 in NESC. This table starts at 1000 V phase to phase. It takes in to account conductor spacings typical on OH systems and that due to these spacings a SLG fault will not or rarely escalate into a 3 phase fault.

The OP was concerned with voltages <1000V and txf less than 125 kVA.

Arc Pro below 1000 V

I spoke w/ support staff @ Kinectrics after I read the last post. They believe their model IS good for voltages below 1000 V.

There is a significant relationship b/t the available voltage and the arc gap. Its the voltage level that is "pushing" the electron flow across the gap, so more voltage is required to push across a larger gap (assuming all other factors are the same). Apparently they have done testing below 1000 V.

I'm not going to get into all of the gory details here, but I would encourage everyone to contact the Kinectrics support staff if they have questions. Their support staff has been responsive, and they seem quite knowledgeable.

They also mentioned that they were hosting some folks from the IEEE 1584 committee today, and that they are engaged in the next round of testing.

Hopefully we will all get to a point in the future where we are asking questions, and getting answers; versus asking questions, and getting even more questions.

See the discussion under the section on software for studies