[TowerTalk] Lightning Protection

[Jim Lux]

On 7/9/13 1:23 PM, Michael Tope wrote:
> On 7/8/2013 12:17 PM, Jim Lux wrote:
>> I think there's an enormous amount of almost correct info about
>> grounding out there. Some of it dates from the days of running a
>> single long wire against a ground rod. Some of it is from electrical
>> codes (which have different objectives). Some of it is just plain
>> misunderstanding of the physics, but has crept into the lore.
>>
>> (that sharp bends have much more inductance than gentle ones...
>> provably wrong, but you see it in lots of places)
>
> Yes, based on what I've heard others claim, I was under the impression
> that sharp bends had a lot more inductance, but it just occurred to me
> that right angle TEM transmission lines work reasonably well even
> without mitering (mitering or curving is better, but transmission lines
> still work without it).
>
> If there was a really large increase in the inductance associated with a
> sharp right angle it doesn't seem like that would be true. I suppose
> sharp bends do concentrate the electric field to a greater degree than
> do smooth bends, so avoiding sharp bends might be relevant in an
> application where there is risk of arcing. Of course, I can imagine a
> lot of scenarios (depending on the exact geometry), where the end-to-end
> length of the ground conductor ends up being shorter if it is curved
> smoothly than it would be if it were bent sharply. In that case the
> inductance would be lower by virtue of the shorter conductor length.
>

The inductance of a turn of whatever radius (as long as it's less than a 
full turn) is roughly the same as the length of the wire involved.  So a 
right angle bend that is 1 foot on a side has roughly the inductance of 
2 feet of wire.  A smooth 1 foot radius quarter circle has roughly the 
inductance of 1.6 feet of wire (the circumference of a quarter circle). 
   Increases in inductance are due to the magnetic field of one part 
interacting with another conductor, and conductors that cross at right 
angles have very little interaction.

As you note, it's the electric field from the sharp corner that's at 
issue here. TIghter bends are more likely to flashover to other stuff.

What you really to avoid is complete circles (no nice tightly curled 
pigtails!), because the inductance goes up as the square.  More to the 
point for lightning, the electromagnetic forces also go up fast.


I didn't figure all this out til I started to really dig into the whole 
"inductance of grounding wires" thing vs the "AC resistance of ground 
wires" and did some FEM calculations which came out (what I thought) 
were weird, and then ground out the analytical formulas (Thanks to 
Vaughn K6ZTA, a co-worker, for helping me on this).  I then ran across 
the original inductance monographs from NBS (they're like the first and 
second ones published back in the twenties, when NBS was first founded).