Say I have an ACB short time delay set to 0.2s that works out to provide a CAT2 rating downstream. This is based on the system parameters which dictate a certain Iarc current through the ACB. However, if the fault current through the ACB was about 30% higher than this calculated value, from the TCC a CAT3 rating would be applicable. Am I right in thinking that, technically, under NFPA70E the CAT2 rating would apply, and technically a further settings reduction would not be required to meet a CAT2 rating downstream.

While of course electrical faults are rare, I think its fair to say it is possible for a fault current at this "30% above" level to occur, but does it need to be considered as an arcing fault to IEEE1584 and NFPA70E?

And yes, I realise the conservative approach would be to just reduce the Iinst so that the entire ACB curve is below the CAT2 curve on the TCC, but I'm trying to work out if this is necessary?

70E does not provide a calculation except to reference others and to provide the tabular method.

It is quite common to see the incident energy increase as you get farther down the line away from an overcurrent device because the line impedance lowers the fault current but also causes the trip time to increase and the net result is often an increase in incident energy. I've also seen the opposite occur.

IEEE 1584 specifies that you calculate the bolted fault current and 85% of the bolted fault current. Both are applied to the incident energy equation but the fault times will be different. Then you take the worst case from both. IEEE 1584 states that this seems to do a reasonable job of modeling the full range of incident energies actually measured in the lab about 95% of the time.

Thanks for your response Paul. I agree NFPA std does not provide a calc, I only included it as it applies the cat ratings.