OK, reality check here. Assume a short run from the box- 25' of 12-2 cable, 50' electrical round-trip to the box. That #12 cable has a resistance of about 2.0 Ohms/1000' and a reactance of about 0.177 Ohms/1000'. Since the PF is close to 1, the overall impedance rounds to 2.0 Ohms/1000'.
50' of #12 wire would have an impedance of 0.1 Ohms. At 120V, that's 1200A (6000% breaker rating) for a dead-short. 1200A is going to cause significant voltage drop, though! If the poco ran 100' of 4/0 Al to the pole pig, voltage drop would be about 40V, and voltage/current through the #12 wires would actually be 80V and 800A (4000% breaker rating).
Approximately 64kW will be dissipated over the length of the wire. The specific heat of copper is 385 J/kg/K. 50' of #12 contains about of copper. It would take 2 tenths of a second for 64kW of power to heat up 1.06kg of copper from 25C to 60C. If the breaker trips within those 2 tenths of a second, there will be absolutely no cable damage whatsoever. And this is conservative- if the cables in the house were longer, current would be lower and they would heat up slower. If a 50 Amp load was connected (2.4 Ohms), and that breaker trips within 5 seconds, there would be no chance of cable damage. Cross referencing the breaker trip curves- surprise! All the numbers you can run very closely match the upper curves for residential breakers.
...but if the breaker doesn't function correctly, any fault capable of tripping the thermal overload of the breaker MAY begin to damage the cable if the temperature rises much past 60C before the breaker finally trips.
[This message has been edited by SteveFehr (edited 11-01-2006).]
