Substation Grounding: Extend grid to cover adjacent cell tower?

[RSChinn]

We have a cell tower in a 15ft X 10ft fenced area with a buried horizontal ground conductor 2ft outside the fence and ground rods connected to it and the fence. This cell tower site has control boxes and equipment that need to be maintained by Ma Bell personnel. This fence is within 15ft of our substation fence and not tied to our substation grid. The WinIGS program shows a steep potential gradient drop of just ouside our fence. The first 30ft of the gradient is very steep at about 100v per ft. Question: Do we tie the corners of the cell tower fence to our substation grounding grid or not? If we do, then we have extended our "equipotential mesa" to include the cell tower area (and the area between the two fences) to protect Ma Bell's personnel during a GPR event. If we connect it, are there other concerns such as "transferred voltages" and if so do we leave their fence separate from our's. The cell tower is on our property.

 

[jghrist]

Are there any communication wires at the cell tower that are at remote earth potential? If so, then transferring the station GPR to the cell tower ground may cause problems. On the other hand, if there is an electrical service at the cell tower, it will be grounded to a distribution neutral that may have the same effect of transferred potentials from the substation ground grid.

 

[RSChinn]

To jghrist: I called the telephone co and they said the cell tower sites still use positrons and even include fiber links as available. The power supply to their site is from our transformer on a pole that is grounded back to our grid via the neutral and is grounded at the pole also. The transformer on some sites may be as far as 200 ft away from our grid. I do not think I have a transferred voltage problem with the signal conductors; the power supply is a "loop" ground going from the grid out 200ft to the transformer and then back 200ft to the adjacent cell site. Is this correct thinking? Now, do I worry about the area between their fence and our fence if the we do not solidly bond their cell site grid to our grid?

 

[Kiribanda]

RSChinn,
Interesting issue.Can you please upload the site plan showing cell tower grid, your grid and transformer location with approximate distances for everybody's use?

 

[waross]

At first thought, won't just bonding the fence transfer your voltage gradient to the equipment inside the fence?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

Since the cell tower ground system is connected to the substation ground grid anyway through the distribution neutral, I think the safest thing to do would be to connect the cell tower perimeter ground wire to the substation ground grid.

Model the connected grounding systems in WinIGS and see if there are any step- or touch-voltage problems.

 

[RSChinn]

Thanks to all. We ran the WinIGS and found that tying the cell tower grid corners to our grid creating an "extended mesh" caused step and touch potentials to not pass very well although the cell site grid was tied via the system neutral already. The cell tower did pass with flying colors in WinIGS when we modeled it with the system neutral connection only and made 5' X 5' square meshes under the cell site. Therefore we are not tying their grid to ours. Your suggestion to run the WinIGS for connecting the two grids was, of course, a good check. Thanks.

 

[jghrist]

The cell tower did pass with flying colors in WinIGS when we modeled it with the system neutral connection only and made 5' X 5' square meshes under the cell site.

Did your analysis include a high side fault at the substation with the tower ground connected to the substation grid via the neutral?

Did the analysis with the grids interconnected include the 5' X 5' square meshes at the cell site?

 

[RSChinn]

To jghrist: Yes, we did a high side fault analysis with the 5 X 5 ft mesh for a grid under the tower site and all passed. Our grid mesh at the substation is 15 X 15 ft and using a 15 X 15 ft mesh under the cell tower failed the touch potential at the cell site area only; the substation passed. If you have other indications that this is wrong, please continue with our dialog. By the way, don't distribution neutrals connected to the grid transfer GPR voltages to the neighborhood house neutrals? I would think this is dangerous, but I guess we live with it. Someone said that some utilities do not connect the distribution neutrals to the substation grid, but simply use earth to get back to the Xo bushing of the substation transformer. If this is true and will prevent transferred voltages, why aren't all utilities doing this?