-----Original Message-----
>From: Jim Brown <jim@audiosystemsgroup.com>
>Sent: Oct 21, 2008 8:49 AM
>To: Tower Talk List <towertalk@contesting.com>
>Subject: Re: [TowerTalk] Using old coax for ground wire
>
>>however NEC bonds are NOT for RF and NOT for
>>lightning dissipation.
>
>I'll take several issues with that. First, NEC bonding to earth
>electrodes IS specifically for lightning protection. That is the
>purpose of an earth connection.
I'll buy lightning protection, but not lightning dissipation (at least not from
a direct strike). As you mention below, it keeps all the pieces at (about) the
same relative voltage, and in the event of a fault or a nearby lightning
induced transient, that AWG17 copper clad will do just fine. The sizes in the
code are apparently chosen for mechanical strength, not electrical properties,
per se.(obviously, it needs to be a conductor, of some sort..) (I'm looking for
the regulatory history of that code article to confirm this.)
>
>One function of NEC bonding between equipment enclosures is to
>provide a path that will blow a fuse or breaker if there is a
>fault that makes the chassis hot. Another function of bonding is
>to cause all parts of the electrical system in a home to rise
>together in the event of a strike, thus minimizing the DIFFERENCE
>between equipment that might be interconnected. It is often this
>difference that blows up equipment. Bonding everything together
>approximately limits that difference to the voltage induced on the
>bonding conductors.
I fully agree. The other fault contemplated in the NEC, particularly with
respect to antennas, is the short from a LV feeder or MV distribution line.
That 14kV MV line that gets hit by the tree in the windstorm and then falls
across your satellite dish or TV antenna or 160 meter beverage.
>
>Second, lightning IS RF. IEEE studies show that most of the energy
>in lightning is in several octaves centered around 1 MHz (that is,
>from about 250 kHz to about 4 MHz).
Yes.. but the NEC bonding rules don't require any special RF-ish treatment
other than "direct routing".. A 200 foot long AWG 17 copperclad would meet the
NEC requirement, and be a terrible RF ground for a vertical antenna (for
instance). In my statement, when I said that NEC grounds aren't necessarily
good RF grounds, I was really referring to use as a ground for an antenna,
where it's expected to be part of the radiating system.
Conversely, good RF grounds are often good lightning dissipation grounds
(although the way the codes are written, you might find yourself adding another
wire, just to meet the specific code requirement).
>
>Third, there is no "RF" reason, other than lightning protection,
>for a connection to earth. A connection to earth does not improve
>the performance of antennas, it does not reduce noise, it does not
>reduce RFI. References saying that it does are WRONG, including
>statements in equipment manuals, ARRL publications, and even
>several questions in the current license exam pool!
How about on a vertical 1/4 wave antenna.. (which was what I was thinking of in
the context of RF grounds).
I concur with the utility (or more likely, non-utility) of grounding for RFI
reasons.
>
>As to the bonding conductors themselves -- the primary problem
>with braid and other small stranded conductors is increased
>CORROSION if they are exposed to the elements. That's why they are
>not recommended for these applications. There's nothing wrong with
>using them if they are not exposed. I use braid for bonding inside
>my shack because it is beefy and more flexible.
Indeed, I suspect that the choice of heavy stranded copper wire for lightning
protection systems is made on the basis of convenience, cost, and durability.
Why carry two spools of wire/braid, when you can use the same thing everywhere.
You pretty much have to use the non-braid outside, and stranded is the only
practical way to use a 4/0 conductor, unless you like bending big copper rods
or bars. If you're installing a air terminal and down conductor down the side
of a building, the advantages of stranded heavy wire over just about anything
else are self-evident.
I guess I should have been more clear on defining what I meant by RF ground..
ah well.
Jim, W6RMK
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|