[TowerTalk] original source of "avoid sharp bends" in lightning

[David Gilbert]

I agree with your reasoning discounting the magnetic coupling effects of 
a bend.  I always assumed that your item 2) was the reason for that 
recommendation, but intuitively it always seemed to me that if the 
impedance between the point of the sharp bend and ground was high enough 
to make the increased electrical field at the bend comparatively high 
enough for a breakover to occur that it wasn't a very good ground path 
to begin with.

73,
Dave  AB7E



On 4/12/2012 6:31 AM, Jim Lux wrote:
> On 4/11/12 10:33 PM, Grant Saviers wrote:
>> might be as simple as:
>>
>> "Don't do anything that increases the inductance of the grounding wire" ?
> the original source might say that, but it's not actually true.  Until
> you get a full 360 loop, the inductance of a bend isn't much more than
> the inductance of the length of wire in the bend.
>
> if you think about it, the increase in inductance in a loop is because
> the magnetic field from one part of the wire interacts with the field
> from another.  Two pieces of wire at right angles don't couple very
> well, parallel pieces do.  A 360 loop has coupling "across" the loop,
> but with an incomplete loop, much of the loop doesn't couple to anything
> else.
>
>
> In any case, that's why I'm looking for the original source.  I suspect
> that it's either:
>
> 1) how things were wired back in Ben Franklin's time, and it's carried
> on to this day because it doesn't hurt, even if it doesn't help, either.
>    (the corollary, it takes more time to explain why not than to just do it)
>
> 2) A sharp bend produces a place where flashover/breakdown is more
> likely to occur.
>
>
> I discovered a few weeks ago that a bunch of the grounding conductor
> guidelines in power substations (something that amateur radio folks are
> unlikely to have to worry about) were based on the behavior of bolts and
> materials commonly used at the turn of the last century.  So that got me
> started..
>
> (that, and Ward's article about 468)
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>