NEC modeling

AM Radio discussion. Directional arrays are FUN!

Re: NEC modeling

Postby R. Fry » Wed Dec 17, 2008 5:43 pm

K9EZ wrote:How would this compare to the 120 radials?

It is slightly better than such a system in pristine condition. Radial systems with 120 buried conductors each about 1/4 physical wavelengths long have an ohm or two of r-f resistance, which means that the groundwave field intensity at a given distance should be several percent less than calculated for this NEC model (other things equal).
Also would there be any difference with 6 elevated radials?

NEC says No. In fact, NEC says that this model even with two or three symmetrically oriented, elevated radials each of the same physical length as in this model still would have the same gain and h-plane circularity as with four, six, or more such elevated radials.

The average r-f current in each of more than four elevated radials will be less than in my NEC model, but that should be the only practical difference.

Also note that using such an elevated radial system will not produce any significant improvement over that of a conventional AM broadcast r-f ground system of 120 buried radials each about 1/4 of a physical wavelength, which is in reasonable electrical condition.

For a given applied power, radiator and r-f ground, once the groundwave signal is "launched" from a given tx site, there is no practical way to improve its distant field strength. That depends on the relative losses along the groundwave propagation path -- which have nothing to do with the radiator, or its r-f ground (buried or elevated).

But, kindly run all of this by your favorite consultant for confirmation before deciding what might be appropriate for a given set of conditions.

RF
R. Fry
 
Posts: 87
Joined: Mon Jun 23, 2008 8:41 am
Location: Illinois

Re: NEC modeling

Postby davedybas » Fri Dec 11, 2009 4:42 pm

I'm modelling a Non-D antenna in a program called 4NEC2...a freebe program that is at a price I can afford. I've used a Non-D antenna in this case just for it's simplicity. However, I'm a bit confused about ground system parameters and thought maybe someone could help me out.

Here goes: The ground system parameters of this program allow for input of "Conductivity" and "Dielectric" values for the earth that you are modelling your antenna upon. Even when you put a 120 radial standard ground into the model you can specify the conductivity and dielectric values of the earth below the radials and outside the radials....

My question is....what value should the "dielectric" be? After all we engineers are familiar with Conductivity values and Conductivity Graphs...but I've never seen any info on soil "dielectric" value. However changing the dielectric values sure does affect the signal strength of the antenna modelled.

Any thoughts...what is common practice?

Dave Dybas
User avatar
davedybas
 
Posts: 164
Joined: Wed Dec 19, 2007 9:16 pm
Location: Buffalo Grove, IL

Re: NEC modeling

Postby Deep Thought » Fri Dec 11, 2009 5:00 pm

The dielectric constant the FCC used to generate the conductivity graphs over soil is 15 and for salt water is 80, as noted at 47 CFR 73.184(a). A more general formula and treatment is described at 47 CFR 73.184(c). For your purposes, ε=15 will do nicely.
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com
User avatar
Deep Thought
 
Posts: 2126
Joined: Thu Mar 20, 2008 10:23 am
Location: La Grange, IL

Re: NEC modeling

Postby davedybas » Fri Dec 11, 2009 9:48 pm

Thanks "Deep Thought"...I should have figured you'd be the one to know the answer to that.

You know as boring as the FCC R&R are ....I never noticed that little bit of info...maybe I should leave a copy in the bathroom book basket };>)

Thanks again!
User avatar
davedybas
 
Posts: 164
Joined: Wed Dec 19, 2007 9:16 pm
Location: Buffalo Grove, IL

Re: NEC modeling

Postby Deep Thought » Fri Dec 11, 2009 10:20 pm

Yer welcome. Certain bits of tech data embedded in the FCC Rules are engraved on the inside of my eyelids, mostly from years of writing computer programs that try to guess what the processing people are going to do on any particular day. That one is from a particularly vexing session involving complex number algebra, a persnickety FORTRAN compiler, and 1980's desktop computer hardware.

I must have stared at those formulas for days, and finally gave up and went to a lookup table. Then, three months after I got my program debugged they re-did the conductivity graphs in the tasty metric system and released the graph data as both tables and Postscript files. I cried.
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com
User avatar
Deep Thought
 
Posts: 2126
Joined: Thu Mar 20, 2008 10:23 am
Location: La Grange, IL

Previous

Return to AM Radio

Who is online

Users browsing this forum: Bing [Bot] and 1 guest