A more dangerous scenario is that given a utility transformer failure, the utility company replaces the transformer with one they have pulled out from another location and is sitting in their yard, the impedance happens to be Z=4.2%. PCO doesn't care, i.e. 2500KVA is 2500KVA.
Using the conventional worst case scenario calculation of infinite bus, the available fault current on the LV side (480V) of the transformer now goes from approx. 63kA to around 67kA. The breaker originally installed rated for 65kA is not rated properly now.
The moral of the story is: if the PCO replaces a transformer to your service, don't think there isn't anything to be checked on the Owner's side.
Food for thought: How many facilities do you think exist that have inappropriately rated breakers, simply because the original designer tried to save a couple bucks and go with the lower kA rated device instead of conservative choice, or they should have used fuses instead of a breaker. When it comes to AIC ratings always error on the conservative side, safety should never be compromised.
Our POCO doesn't ever give us the actual Z of the transformer, but instead they give us their spec. for the minimum Z any of their transformers of a certain size will have.
But then, there isn't anything to stop them from replacing a 300 KVA with a 500 KVA, so the same problem is still there, just less frequent.
But if you do calcs with infinite primary, that usually gives a pretty good buffer between the calculated and actual available fault current.
Steve