[TowerTalk] Grounding of cables to tower?

jimlux jimlux at earthlink.net
Wed Oct 18 20:41:21 EDT 2017


On 10/18/17 4:21 PM, Wes Stewart wrote:
> On 10/18/2017 3:34 PM, Jim Brown wrote:
>> On 10/18/2017 1:47 PM, Wes Stewart wrote:
>>> A couple of questions:
>>>
>>> 1) on page 15 of your referenced document you say a distant tower 
>>> should not be bonded to house (shack) ground and go on to connect the 
>>> coax shield.  How is that not a connection between the two?
>>
>> It is a connection at DC. But at RF, the inductance of that connection 
>> dominates, so it's not a bond --  a bond must be a low impedance 
>> connection.   :)
> 
> The NEC says of bonding in Article 100: "The permanent joining of metal 
> parts together to form an electrically conductive path that has the 
> capacity to conduct safely any fault current likely to be imposed on 
> it."  By this definition, the coax shield is a bonding conductor.  
> You're making stuff up.

Ah, the fine problem of the meaning of words in different contexts - in 
Electrical Code land - bonding could well be done by the coax shield, 
with a couple potential problems:
1) current carrying capacity - RG174 probably wouldn't hack it. I don't 
know about RG-8/RG-213 sized.
1a) You'd have to make some reasonable case about what a plausible fault 
current might be - this is the beauty of Class 2 Energy Limited devices 
(your doorbell)- the maximum fault current is quite low, so almost 
anything works. If you're running coax to something also supplied by an 
AC power cord, it's a bit trickier. Probably would depend on the fusing 
or overcurrent protection of the power supply.

2) In general anything with a connector doesn't count as "permanent 
joining" - if you strip the outer jacket and install a connecting clamp 
around the shield, that would be fine (and, in fact, is recognized in 
the 800 section articles were coax features in other ways).  But if you 
have a bulkhead connector - it's not "bonded" as far as NEC is concerned.
2a) Your local inspector makes the final call, they may be totally cool 
with a connector, especially if it were staked in some way, so it's not 
trivially disconnectable.  After all, they're perfectly happy with screw 
terminals for bonding connection of the "grounding conductor" (aka green 
wire ground)

However, in the RFI world, bonding has an entirely different meaning.

And in the lightning protection world (another NFPA standard), again, 
bonding means something different.


Hey, at least the NEC is getting rid of the "grounding" word which is 
just creating total confusion because you talk about the "grounded 
conductor" (neutral) and the "grounding conductor" (green wire), all of 
which gets very confusing when you have systems which are deliberately 
isolated from earth ground





>>

>>
>> Another point that I should have made is that if mains power is fed to 
>> the tower, the green wire must, by code, be run with the phase and 
>> neutral conductors, and it should be bonded to ground at the tower. 
>> The reason is what Jim Lux articulated earlier in this thread (or 
>> maybe another thread) -- the bond must be there so that a fuse or 
>> breaker blows in the case of a fault from phase to ground.


I'm not sure the green wire has to be bonded to ground at the tower - it 
can serve its purpose without it (short from line to case blows fuse).

It's probably one of those "accessory structure" kind of things that 
make code interpretation "fun" - does a subpanel in a workshop that is 
physically separate from the house have to have a separate green wire 
grounding system?  Does the neutral have to be bonded to ground?

And to get exciting, we can talk about the differences between the 
grounding/bonding requirements for portable and stationary generators 
used for emergency power.


>>
>> Thanks for catching the conflict -- I'll have to change that when I 
>> have time.
>>
>> 73, Jim K9YC
> Wes  N7WS
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk



More information about the TowerTalk mailing list