Electric fences

[AM9]

It's not plain DC. It's a very short duration pulse. So the plain resistive potential divider approach is not adequate; you have to consider the load (fence wire and short to ground) as (most likely) a transmission line. Without going into technical detail the point is that the voltage does not rise simultaneously all the way along the wire; instead the voltage pulse travels at a particular speed, and if it gets to you before it gets to the short to ground, it doesn't know it's been shorted to ground yet, and it can still find you.

The dangerousness of the pulse is limited not by having an especially high source impedance, but by the amount of energy delivered to each pulse. This is not only a more dependable method, it pretty much "comes for free" without calling for any additional parts or any additional design effort.

The propogation speed of an electic pulse along a wire through air is likey to be around 0.5 to 0.9 c, so even if the wire is long enough to have much of an inductance, it will have little capacitance. Also, practical fences of the type being discussed here are rarely more than 500m long per generator feeding it. Assuming the most inductive viable line, the largest difference of a pulse's time of arrival at any two points would be well under 2us. From advertisements, I see that pulse lengths are claimed to be around 100us, and to make the pulse safe, the total energy would be limited. So if the pulses are nominally square, then at least 98% of that safe level of energy in the pulse would be shunted to ground even if the short was on the opposite side to the generatror.

 

[Annetts key]

The propogation speed of an electic pulse along a wire through air is likey to be around 0.5 to 0.9 c, so even if the wire is long enough to have much of an inductance, it will have little capacitance. Also, practical fences of the type being discussed here are rarely more than 500m long per generator feeding it. Assuming the most inductive viable line, the largest difference of a pulse's time of arrival at any two points would be well under 2us. From advertisements, I see that pulse lengths are claimed to be around 100us, and to make the pulse safe, the total energy would be limited. So if the pulses are nominally square, then at least 98% of that safe level of energy in the pulse would be shunted to ground even if the short was on the opposite side to the generatror.

I suspect that the output is being driven by a pulse transformer. But I’ve never had the chance to pull one to pieces…