Directional EF and cable networks 1

[veritas]

I have a 22kV extensive cable network fed from a 22kV bus via several outgoing feeders. The current EF settings is 6A, 0.5s on all the feeders. System is resistively earthed limiting the EF current to 25A. If there is a fault on A-phase of feeder A, will there be an infeed from the A-phase on the adjacent feeders into the fault? If so will it necessitate directional protection for the feeders?

 

[davidbeach]

If multiple outgoing breakers are sharing the fault current, directional won't help as it will be forward for all of them. And if more than four feeders, then the split will be less than 6A and none of them will trip.

Somehow, I think that your description and my picture of it don't actually go together. Perhaps a one-line diagram could help.

 

[veritas]

I've uploaded a simple SLD to illustrate the fault I'm talking about. Fault is on Feeder A. Fdrs B, C and D contribute currents IB, IC and ID to the fault. This is my first question - the way I see it the cable capacitance has "stored voltage". This tends to support the faulted phase voltage - how exactly I am not sure. Compare this to inductive energy stored in a magnetic field like an induction motor, which discharges current into a fault.

Let's say that the cable capacitance does produce current IB, IC and ID contributions to the fault. I see no need for directional EF protection on these feeders for several reasons.

1) Duration of IB, IC and ID will be typically < 5 cycles.
2) These currents are literally at the threshold of the zero torque line, in fact just outside it as the zero torque line is usually concave. But they could lie just within the operate zone due to resistance resulting in unwanted operation.

 

[Marmite]

Are all four of these feeders operated in parallel?
If not the only way I can see feeders B,C & D affecting the earth fault current seen by the relay on feeder A is if they have a coincident earth fault, for example a cross country earth fault. Any stored charge in the cable capacitance or system inductance creates a DC current at the instant of fault.

Regards
Marmite

 

[davidbeach]

So, what is the source for the contribution from the adjacent feeders? Cable capacitance isn't a source.

 

[veritas]

The energy stored energy in the cable capacitance would be the source. This is my take anyway. As I said, I expect this to last only a few cycles. Also, feeders are not in parallel.

 

[Marmite]

Energy would transfer from the system to the cable capacitance and be stored as a charge on the cable in the first quarter cycle, and that energy is returned to the system over the second quarter cycle. Any trapped charge would dissipate into a fault very rapidly as a DC current. Fraction of a cycle rather than a few cycles. Too fast to affect overcurrent protection, even if CT's could transform it accurately. This is a transient DC current, not fault current.
Regards
Marmite