Fault current

[BCATIST]

Recently we had an incident involving a falling branch on an LV power line which energised a colour-bond fence.

Our (protection) fuses are generally set to operate above specific load currents in order to facilitate overload tripping. Faults involving vegetation are often high resistive earth faults that are accompanied by arcing. In this case the fault current was considerably lower than the normal system load current. As a result, the protection system was unable to pick up low residual earth fault current. In this instance the fallen power lines came to ground and remained live constituting a major hazard for the general public.

I am just a tad confused as to why the current level was considerably lower than the normal system load current if it was a short circuit fault and the resistance to ground is reduced due to the conductivity of the colourbond fence (even more if the fence is earthed).

 

[magoo2]

The fence may provide a low impedance (resistance) to ground but you also have contact resistance between the LV line and the branch and between the branch and the fence. Either or both of these can be a high resistance.

 

[BCATIST]

My point is that wouldn't the fault current be higher due to the reduced resistance to ground given the conductivity of the fence?
Yet in this case the current level was below the pickup of the fuse. A sustained vegetation ground fault should have a greater chance at reducing the current level below the pickup prior to any arcing ?

 

[davidbeach]

Earth itself can be a very poor conductor. This is a problem that has eluded solution for a very long time. There are some high end feeder relays that use high frequency signature analysis to attempt to detect high impedance arcing faults. These aren't yet widely used within the industry.

 

[cranky108]

Another issue you may not realise (or maybe you do) is that a fuse may not blow at a current less than twice the rating of the fuse.
Or said another way, a 25 amp fuse may not blow at less than 50 amps.

A tree branch fault that does not clear is a common problem. It dosen't happen every day, but it is common.

Fuses don't offer a lower neutral current protection, but they are reliable and cost effective. You may want to investigate if you need a lower neutral current protection, and if your load can stay within the limitations of a lower neutral current protective device.

 

[bacon4life]

Paint and polymer coatings tend not to provide good conductive paths. From the installation diagram for Colorbond, the fence seems made mostly of slip together channels. Polyurethane has a dielectric strength of 24 kV/mm, so it doesn't take much paint thickness to provide temporary isolation. A metal fence near electricity is kind of like Schrödinger's cat: the fence is simultaneously both insulative and conductive, but you don't know which one until it kills you when you touch it .

For intentionally grounding fences near electrical substation, the paint is always removed to allow tight metal to metal contact. Substation fencing installers pay careful attention to make sure there every piece of conductive material is bonded together to ensure the fence is tied solidly to the earth.

 

[jghrist]

I assume that the fault path was through a branch that was contacting the LV line and the fence. You have several sources of high resistance that would reduce the fault current:
[ol 1]
[li]Contact resistance from the branch to the line.
[/li][li]Resistance of the branch.
[/li][li]Contact resistance from the branch to the fence.
[/li][li]Resistance of the fence coating at the branch.
[/li][li]Resistance of the fence coating at the fence post.
[/li][li]Resistance of the fence post to earth.
[/li]
[/ol]