[TowerTalk] Grounding of cables to tower? (N3AE)

[jimlux]

On 10/18/17 7:11 PM, Shawn Donley wrote:
> Interesting discussion.  My tower is10 ft from my detached garage which is about 100 ft from the house.  The garage has an underground power feeder from the house service panel which is about 230 ft away on the opposite side of the house.  I had to have an electrical permit and inspection thanks to the 800 section of NEC.   The inspector insisted on two things:
> 
> 
> 1.  The ground and neutral in the garage sub-panel could NOT be tied together in that sub-panel.  Grounding conductor in the feeder bonds the sub-panel to the house service panel.


That's typical - neutral (ground*ed* conductor) is bonded to ground 
(ground*ing* conductor) at only one point.


> 
> 
> 2.  The tower had to be bonded to the ground bar in the garage sub-panel.


That too makes sense - bonding to the closest thing that's in the 
"grounding system".



> 
> 
> The garage sub-panel Is bonded to a ground rod just outside the garage and I bonded the tower to that ground rod with solid #4, which satisfied the inspector.   The feeder from the house does include a grounding conductor which is bonded to the garage sub-panel as well as the house service panel grounding bar.

The code has no problem with multiple "earth connections" - the 
"grounding" system can have many connections to ground.


> 
> 
> The tower also has a radial grounding system with (currently) about seven 8 ft ground rods plus a Ufer using the base rebar bonded to the tower anchor bolts.
> 
> 
> So...my tower is bonded to my electrical service “ground” which is 235 ft away from the tower, and through a wire which partially runs through the house basement before going underground.   At RF frequencies, I doubt that the tower is really bonded to the house service panel....DC yes, RF no.  Not saying this is the proper or best technical solution but it appears to be the requirement around here to pass electrical inspection for a tower.
> 
> 





> I’ve debated running a separate bare #4 or #2 underground from the tower to the house service panel with a few ground rods along the way, but it seems to me that would create a big loop that could inductively pick up currents from a nearby strike and only make things worse.
> 


And therein lies the challenge - code/good practice focuses on fault 
current handling and not starting a fire or killing people. It cares not 
for RF properties.

One of the classic ways to solve the "giant ground loop" problem is to 
provide galvanic isolation - a transformer.

In your case, one could put a 1:1 transformer at the remote building, 
primary side fed from the house, secondary side feeding your subpanel, 
where neutral and ground would be connected(because it's an entirely 
separate system).
There would be no "ground path" from house to garage.

A 15 kVA "dry transformer" is about $500

Of course, if you run your coax, bonded to the tower, bonded to the 
garage grounding system, back to the shack in your house, you've now 
created a path from "grounding system in the garage" to "grounding 
system in house", but that's the only path...




> 
> I wrote about this a couple of years ago in TowerTalk but the thread quickly re-focused only on #1 above.
> 
> 
> N3AE
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>