Calculating Attenuation of GPR Transfer on OHL Grounds/Buried Grounds

[Kayaker11]

I have read in reports and seen during testing, that substation GPR attenuates as it is transferred out to remote areas that are directly connected (via intentional grounds) to remote facilities. In the testing I witnessed, the substation had a ground fault imposed at a non-60Hz frequency. Step/touch and GPR were measured at a nearby generating system where the gen grid was tied to the substation grid with a couple buried 250MCM bare conductors. GPR was seen to attenuate from that being imposed on the substation grid.

In the case of the reports, the direct connection was over the ground wire on a overhead line out to a mining pit. I have a great deal of faith in the engineer running the tests and writing the reports but he is long since retired to ask questions of. My question is, has anyone ever seen a method for calculating the attenuation of GPR transfer over various types of ground connections?

I am not even certain I understand the mechanism by which it attenuates (capacitive charging current?) to begin coming up with a way to calculate it. The Canadian mining code requires that we assume that the full magnitude of substation GPR is transferred out into mining areas if they are bonded together. In older mines where the distribution system is such that it is impossible to separate the substation ground from the mine ground system, and the economics of an isolation transformer are unreasonable, a method that might prove that GPR transfer levels are safe would eliminate a lot of heartache.

Thanks in advance for your thoughts on this.

 

[Gr8blu]

Rule #1 - never assume that the other person understands what you mean when you use an acronym or other abbreviation. For example, I suspect you are talking about Ground Potential Rise and not Ground Penetrating Radar or Gross Profit Ratio ... but you never know.

For your "test" case, you had a buried pair of 250 MCM conductors between the generation and substation grids. Highly likely that a conductor of that size is not a solid, but a stranded construction - which means it is not a purely resistive element. That in turn means the observed impedance of the conductor can be affected by frequency - and obviously, by length.

Converting energy to motion for more than half a century

 

[jghrist]

If the ground connection between the main and remote ground grids is by overhead lines or insulated cables, you could treat the remote ground as a terminal and use software such as SES FCDIST to determine how much current flows to the remote grid and the GPR at the remote terminal.

If the connection between the two grids is by bare conductors, then you would need software that considers the voltage drop in the grid conductors. SES MALT considers all the ground conductors to be at the same potential. SES MALTZ calculates the voltage drop in the ground grid wires. MALZ calculates GPR at different points on the grid.

 

[Kayaker11]

Ah, I think I see how I'm looking at this incorrectly. I was thinking of this as a single substation grid that gets a GPR and transfers that full potential onto a connected wire that is a "floating", carrying off full GPR out to the remote areas. But if I look at is a second remotely connected grid (because there is grounding at the end of this wire, just really hard to model), I can use a current split factor into the remote "grid" of ufer grounds and buried conductors.

Thanks, I will look at modelling options.

 

[jghrist]

the gen grid was tied to the substation grid with a couple buried 250MCM bare conductors.

In this case, you can't just treat it as a second remotely connected grid with a current split factor. The buried bare 250 kcmil conductors will dissipate current into the earth. If the distance between grids is large, then there will be a voltage drop and a reduction in GPR, but it won't be easy to calculate. You need a program like SES MALZ that account for voltage drop in the buried wires.