Here is the quiz:

The IEEE 1584 standard is fairly explicit on the limitations of its incident energy calculation model.
If I use the formula to find the arcing current of a system under 1000V (P.10 Std 1584-2002) I cannot go under a bolted fault current of 700A.

Is that mean that in my system, at 600VAC and a bolted fault current of only 0.200kA, there is no arc that can be sustained? I doubt so. I am pretty sure there is enough energy there to sustain an arc.

Is there an other model for a bolted fault current under 700A?
Or should I consider that no arc would be sustained at that low current?

The 700 Amps (and 106 kA) are just the ranges where the model is considered valid. Outside those ranges, who knows. So to answer your question, I see two options:

1) Default to the theoretical Lee equation
2) Use the IEEE 1584 model recognizing that it is outside the range of validity.

Neither option is a good one but those are the only options available. I'm sure you will find the incident energy will be minimal.

I'm don't know if you can sustain an arc at 200 amps easily at 600V however, given the right conditions, anything is possible.

Thanks Jim,

You are right, with the good conditions, everything is possible.

I am guessing its more so to do with the incident energy or the impact of the arc (even if sustainable) is negligible at the assumed fault condition.

It clearly is "possible" to sustain an arc at low currents (and low voltages) because it is done every day. That is all arc welding with an AC "buzz box" type arc-welder is. OTOH it also demonstrates how unlikley it is that the arc will be maintained at those levels, since it requires a trained operator to start and maintain the arc for any length of time.

Thank you gents,

I understand the variability and the sustainability of an arc at this low voltage and current is not clearly defined in arc flash documents...

I saw from our CSA Standard... that at some point, and in my situation, it is possible to consider a maximum of 2 seconds of exposition even if the clearing time is theoretically longer, since the operator is most likely to get out of there. It would not be possible to do so if the operator was in a nacelle truck or something but in a non-confined area our code allows this approximation.

All that comes from the fact that my faulted bus is at the secondary of a transformer, with no secondary breaker. The fault current reporter at the primary is then way smaller and take longer to the protection to clear.

Thank you again for your input on that issue gentlemen (the masculine here include you too Ladies) it was very helpful.