[TowerTalk] Earth Ground: How good is good enough

[Lux, Jim]

On 9/9/21 2:51 PM, Artek Manuals wrote:
> Jim et all
>
> So I will say it AGAIN. What I want to do is measure the ground 
> resistance of a multi ground rod system in VERY POOR soils in my case. 
> WHY I want to know this number or what I intend to use it for IS NOT 
> relevant to the discussion at this point. Please don't get your cart 
> before my horse.
>
> I will happily spend the $10 if N0AX's book ...IF...it talks about 
> actual measurement techniques for deriving the resistance of a ground 
> system composed of several ground rods at any given frequency. Those 
> of you who recommended the book please verify for me that it talks 
> about actual measurement techniques WHY I want to know is not relevant 
> to the question.
>
> What I want to do is to be able to describe the problem with numbers 
> using verifiable measurement techniques at different frequencies. 

Let's start with the basics - what frequency range are you interested in?

For sort of "raw starting model" the IEEE standard on grounding IEEE Std 
142 - Chapter 4 has equations for various combinations and shapes of 
grounding electrodes that take into account their mutual interaction.  
Other chapters describe measurement techniques.

In general, though, this will not take into account RF - that is, it's a 
quasistatic model for DC or AC where the frequency is low enough that 
everything is a "tiny" fraction of a wavelength.

If you need to take into account  RF effects, I'd probably start with 
NEC 4, which handles buried wires - of course, it assumes "uniform 
ground".  you could model each ground independently, and the 
interconnection network.  People have used NEC for modeling lightning 
strikes, and the flow of current in the ground network, for instance. 
Google for papers on "lightning numerical electromagnetics code" 
("lightning NEC" just gets lots of references to the electrical code)

If your soil is not homogeneous, then it gets much more interesting.  
Probably best modeled by some sort of Finite Element 3D code - You build 
a 3d array of little voxels, each with the EM properties, and then solve 
for the fields.  I don't know of any off the shelf tools that would do 
this - I've seen plenty of "home grown" models - Modeling in a 
non-homogenous medium is pretty standard for the tools, but they usually 
are looking at propagating a wave or impulse, not actually shoving a 
conductor into the medium.

There are plenty of books on computational electromagnetics, but I'm not 
sure I could recommend any that I have - they're more on "basic 
algorithms" and analytical approaches, and computational efficiency - 
that is, one could, in theory, write software to model it after reading 
the book (hah - you'd better have a lot of other experience too).


You might look at the geophysical prospecting literature and books - 
they do a lot of ground resistance measurements at various frequencies 
(up to the low MHz range) with wildly varying electrode arrays.  The 
other thing sort of related is geophysical resistance tomography - where 
they try and determine the existence or shape of buried 
buildings/artifacts, as well as ore bodies (depending on the scale)