[TowerTalk] Radials Questions - 270 or 360 degrees.

[Jim Lux]

At 04:08 PM 1/5/2005, Bill Coleman wrote:

>On Jan 5, 2005, at 11:50 AM, Red wrote:
>
>>EZNEC modeling indicates a very slight assymetry and reduction of gain 
>>resulting from omitting 90 degrees segment of the radials.  Both figures 
>>are within the inaccuracies associated with errors and assumptions in the 
>>modeling.
>
>Is 1.1 dB considered very slight?


1.1 dB is very slight in the context of using EZNEC to model ground planes, 
something it is not particularly good at (especially if the wires are 
buried, where it probably doesn't work at all).  The error in the model is 
probably a lot bigger, as Red said. Upshot is, don't trust the 1.1dB 
number.  You could have 1 dB more gain, or 1 dB less gain, or no change at 
all.



>>The assymetry, 1.1 dB, can be reduced to 0.2 dB by removing the opposing 
>>90 degrees of radials, at the cost of about 0.5 dB of maximum gain.
>
>This doesn't seem like a good idea, then. The net change would be to 
>increase the gain in the assymmetric direction by 0.3 dB, at the cost of 
>0.5 dB elsewhere in the circle....

I would maintain that this change is insignificant compared to the other 
uncertainties in the model.

Modeling buried radials is exceedingly non-trivial.  Even with a code like 
NEC4, which does handle buried wires correctly, would need a good model of 
your soil properties, with a resolution comparable to the other stuff 
you're modeling.

Somehow, I doubt any of the readily available inexpensive modeling codes 
would adequately model the effects of your house, which will dominate any 
directive effects of your asymmetrical radial field.


>>The performance will benefit some from using more radials, even if, in 
>>the interest of limiting the amount of wire used, some are shorter than 
>>60 feet.  However, variables associated with modeling errors, assumptions 
>>of ground characteristics, terrain features, wires and pipes in the 
>>structure, etc. are generally greater than the calculated differences in 
>>performance.  Put down as many radials as you can and don't worry much 
>>about the 90 degree gap.  If you can fill the gap without introducing 
>>excess interference or complexity, do it.


There you go.. the real useful advice.. put down as many radials as you 
have time/money/patience for. Don't worry about the length (well... 
extending past, say, 1/2 wavelength might be a diminishing returns)


>Well, that about answers my first question -- how about the other three 
>questions?
>
>>Question 2: In looking at option B, is the vertical segment going to have 
>>an adverse effect on the pattern in the NW to NE direction? Is the 
>>vertical segment going to change the phase of the currents and ultimately 
>>destroy any gains by adding the radials?

No way to predict. One would probably be safe in saying that there would be 
"some change", but it might not be detectable.  That wire is presumably 
next to something (earth, concrete, interior structural members, etc.  All 
of that will have an effect, and one that is basically impossible to 
predict (at least without spending a lot more time and money).


>>Question 3: Is there any coupling or danger associated with having these 
>>radials inside the house? Is there a potential for high voltages to 
>>appear, or to have the elements radiating RF inside the structure? Is 
>>this different from option A or B?

Yes.. and this is probably the best reason to NOT do it. Putting antenna 
components (of any kind, connected anywhere in the system, ground radial, 
radiating element, whatever) in close proximity to people and/or flammable 
stuff is not a good idea.  If nothing else, since you couldn't do a 
credible model, and I assume you intend to use it for transmitting more 
than a trivial amount of power (e.g. a few milliwatts), you'd have real 
trouble complying with the FCC RF safety regulations (47 CFR 97.13). How 
would you assure that you're not exceeding the RF exposure limits?  (And, 
no, I don't think the categorical power exemption for amateur stations 
would apply here.  Take a look at page 9 of OET Bulletin 65, Supplement 
B... it's on the web)


Jim, W6RMK