Maximum bus fault

[PeterVenkman]

I know that the maximum fault current flows in a line can be during a contingency due to mutual inductance. That is for a line but can the maximum fault current seen at a bus ever be during a contingency or does it always decrease from the base case with any contingency?

 

[davidbeach]

Yes, there can be contingencies that can increase bus fault current. For instance I've seen cases where 230kV faults go up when a 500/230kV transformer connected to the faulted bus clears. Seems that the whole 500kV voltage gets pulled down and remote sources supply less fault current then they do when the 500kV voltages are higher. Not in every case by any means, but there are weird system configurations that can do most anything.

"Never" and "Always" are terms best avoided when doing power system analysis.

 

[odlanor]

davidbeach
You are talking about contingencies that can increase bus fault current. , but in your example the contingencies reduce bus fault current. I did not understand.

 

[davidbeach]

I'm not coming up with any specific examples at the moment, but I know that during prior fault studies I have found areas in our system where there are single contingencies that have increased bus fault current. Took a bunch of work to determine that the results were correct. Has to do with the contingency allowing the overlaying system voltage to increase significantly making for more fault current into the bus from more remote sources (that have less affect on the higher voltage system) and that increase in current makes up for the loss of fault source taken out by the contingency. The best I'm coming up with at the moment has the removal of a 6500A source reducing the bus fault by only 1000A.

The cases may be very rare, but I'm pretty sure I'm remembering that it happened because it caused me so much consternation. A system configuration change since then (relocation of a bulk transformer) may have removed that possibility. Made coming up with relay settings a real bear when system normal isn't the worst case.