No Subject

          Tuesday, 24 January 1995 10:40:46
          rec.radio.amateur.anten Item
  From:           root-at-ancc-dot-com,News Only
  Subject:        Re: Best way to install ground rods?

>I have considered jetting down a ground rod but I don't think the contatc
>with the ground would be as good as a driven rod.

I have heard exactly this.  In fact, the ARRL made mention of it in a 
past issue, that in many types of soils, most of what is left after using 
water pressure to make the hole, is stones.  All of the conductive earth
is washed away, and the worse conductors are left in contact with the 
ground pipe.  Measured ground conductivity is worse after using the water
method.  The other thing is that pipe tends to clog, then freeze, and 
split open.  I much prefer the 5/8" galvanized ground rods that are 
commercially sold.  Yes, you have to beat on them for 10 minutes to get
one in, but unlike pipe, they don't buckle under the beating, and they 

Bruce N9EHA

Remember that a lightning bolt has a surge current of around
8,000 amperes. You have to size and bond your conductors to
handle this load. The duration of the pulse is short, however,
only about 20 coulombs of charge is exchanged in the typical
strike, so the conductors don't have to be large enough to
handle the surge *continously*. Copper conductors equivalent
to #6 solid wire are sufficient. 

Lightning is RF, though most of its energy falls below 2 MHz, 
so the skin effect must be considered. That's why solid strap 
is preferred over round wire. Strap has a larger surface area, 
pound for pound, than round copper wires. Copper pipe can also
be used, but it's surface area will be half that of copper
strap with a width equal to the pipe circumfrence. That's 
because the *inner* surface of the pipe forms a waveguide 
beyond cutoff for the lightning RF currents, and isn't effective
at carrying away the surge. Strap, on the other hand, can fully
exploit *both* surfaces. (Pipe *does* have somewhat lower inductance,
so there is a tradeoff here.)

Woven braid conductors should be avoided for grounding runs because
braid has about 5 times the impedance of smooth solid strap on a
pound for pound of copper basis. There are a couple of reasons
for this. First, the braid strands weave in and out, adding inductance,
and second, because the skin effect tries to force currents to the
surface, while the individual strands keep diving into the middle
of the bundle, the currents try to flow from strand to strand along
the outside of the braid. Since the mechanical connections of one
strand to another are fairly loose, a high resistance path is formed.
Fully *tinned* braid is better, since the strands are now bonded 
to each other better. However, solder *will* melt during a strike,
and you should plan to depend only on mechanically bonded connections,
IE clamps or cadwelding.

Since you are building your house, you have an opportunity now to
install a *Ufer ground*. Mr. Ufer developed this technique during
WWII for army ammo bunkers. The NEC approves its use for commercial
and home grounding systems. In essence, a Ufer ground is just rebar
in concrete. When the builder is preparing to pour your slab, make
sure all the rebar in the slab is bonded together, either cadwelded
or mechanically clamped, before the pour. And make sure to leave a
convienent attachment point exposed. A rule of thumb for a Ufer
ground is that it takes about 20 feet of 0.5 inch rebar to absorb
8,000 amperes of surge. More is better. The rebar should be embedded
in at least 4 inches of concrete.

The way a Ufer ground works is through two paths. First it forms
a large capacitance to Earth. This is an excellent RF coupling.
Second, concrete's ions generally are more conductive than native
soils, so you have a large number of virtual resistors in parallel
connected to Earth that offer a lower resistance than would a
smaller collection of driven rods. Earth is actually a lousy 
conductor. Most currents are dissipated through Earth by capacitive
coupling and arcing from soil grain to soil grain. Concrete is a 
better conductor since the grains are tightly bound together.

Gary Coffman KE4ZV          |    You make it,     | gatech!wa4mei!ke4zv!gary
Destructive Testing Systems |    we break it.     | emory!kd4nc!ke4zv!gary 
534 Shannon Way             |    Guaranteed!      | gary-at-ke4zv.atl.ga.us
Lawrenceville, GA 30244     |                     | 

-- Mark
       _/_/_/   _/_/_/_/       Mark Conway
      _/    _/    _/          Deep Thought BBS, Auckland, New Zealand
     _/    _/    _/          A FirstClass(tm) Macintosh GUI BBS
    _/_/_/      _/          Internet: mconway-at-deepthnk.kiwi.gen.nz