[TowerTalk] newbie question on inverted-L's, tuners, and radials

[hasan schiers]

Hi Eugene,

My comments are interspersed in your posting. I run an inverted L for 80m 
with full radial field and I LOVE the DX Engineering radial plate...first 
rate, stainless steel hardware...very nicely done, with clamping available 
for a ground mast ...it's nicely done.

First of all, if you run a long wire, without any consideration for 
resonance and feed it as an inverted L, you have several things to consider.

1. If you go straight up for 40' and then straight out or sloped out for the 
remainder, you can have a VERY good inverted L for 80 meters, assuming a 
total length of 67'. However, that same length on 40m will be a half wave 
vertical and will present a VERY high impedance (2000 ohms or greater) at 
the base that your tuner could have a really bad problem with. It would do a 
good job of radiating (very efficient), if you could match it properly.

2. Considering it as an inverted L, as the vertical portion gets longer 
witih respect to wavelength, the radiation angle or take-off-angle will 
start to tilt up ...when you get much higher than 5/8 wave for the vertical 
section, you will lose most of your DX performance. In the case of a 67' 
total length wire, with 40' being vertical, you will get excellent dx 
performance (assuming a decent radial system and acceptable soil conditions 
in your area) on 80, 40 and 30 meters. Once you approach .75 wavelength, the 
pattern really starts going up severely.

To avoid endless speculation:

Download a free copy of ELNEC from Roy Lewallen's site and model the antenna 
...change the freq and you will see the effect of having the wire too long, 
and you will also see the effect of freq on the feedpoint impedance.


----- Original Message ----- 
From: "Eugene Hertz" <ehertz at tcaf.org>
To: <towertalk at contesting.com>
Sent: Wednesday, June 21, 2006 1:40 PM
Subject: [TowerTalk] newbie question on inverted-L's, tuners, and radials


> Hello all,
>
>
> Feedline comes from my second floor shack down to the ground.  I will 
> (hopefully) use a remote antenna coupler, a Harris RF-601A.  This tuner 
> will be grounded using radials (see next paragraph).  From the tuner, a 
> single wire will emanate up the side of the house to a mast on my chimney. 
> This should get the height of the vertical portion of the antenna about 
> 40' up. The wire will then traverse across my front lawn to a tree about 
> 140' away and still about 40' up.  I would love to be able to get some 
> signal out on 10-160.
>

A non-conductive mast would be best. 180' total length is going to produce 
some "interesting" impedances for your tuner to match as you change bands. I 
would be most concerned about poor performance on 20m and above, due to the 
vertical element becoming too long. Again, back to modeling the antenna as 
you describe it with ELNEC.

>
> Here come the questions:
> 1. Most of the discussion I read about radials had to do with "vertical" 
> antennas.  I am wondering if all those discussions could be more 
> generalized to pertain to all end-fed antennas such as the "L"? Is there 
> anything different about the use of radials (sizes, locations, numbers) as 
> applied to L's vs verticals?

No...there a VERY important differences. Horizontal vs Vertical antennas and 
then the Hybrid you are proposing. That's why I refer you to ELNEC so you 
can SEE what is happening.

>
> 2. I have an issue with the possible location of my radials. The vertical 
> part of the antenna is very close to my house, therefore My radials can be 
> 180deg _away_ from the direction of the horizontal wire, as the wire 
> travels over the house (not exactly, but you get the idea).  In fact, I 
> may only have some short radials 180 deg away from the antenna, but can do 
> longer ones +/- 90 deg from the horizontal wire.  Question is, with a long 
> horizontal component, is there any necessity to have the radials "beneath" 
> the horizontal portion? For example, if I can get 30 radials in the ground 
> but none of them are beneath the antenna is this terribly worse than 30 in 
> the ground where some are beneath the antenna?

You don't need any radials "specifically" for the horizontal portion of the 
antenna. You want your radials to run outward from the base at the feedpoint 
and to take up as much "area" as you can. So if you have 180 degrees (half 
of a circle) available for radials, put your 30 radials in the half-circle 
and forget about it. Is it ideal? No. Will it work well anyway, most 
probably yes.
>
> 3. With the heights and lengths described above, should I think about 
> shortening/lengthening the wire (perhaps to get onto 160?) Or should I let 
> the tuner do the work and just make the longest wire possible? Bands of 
> interest are primarily 160,80,40,20,15,10 (until I get my urt-23 running 
> then its all ham bands).

If you model this antenna at a specified length you will see that at some 
point it becomes too long to be useful, or its characteristics change in  a 
way you do not like. For certain freqs, longer is NOT better. For certain 
freqs, your tuner won't be able to match the length you have chosen. Getting 
it to work on 160, may make it terrible on 40m and above.

Try this idea:

Pick your "ideal" bottom band (my guess would be between 80 and 40m). Design 
it as a resonant inverted L for that band, and brute force tune it for the 
other bands you want to work  Trying to get it all by making it long is a 
poor solution. If you design it for 80 it will work well for 80/40/30, and 
probably ok for 20. You might surprise yourself with 160m. with this same 
design.

If, on the other hand you design it for 160, then you will get good 160/80/ 
40 probably ok, above that, it's too long for a good solution. AGAIN...go 
back to ELNEC and try various lengths and you will see that you have some 
choices to make, there is no free lunch. Longer is not necessarily better, 
unless you really just want a random length wire antenna...and its 
performance will be...well...kinda random...all over the place.

>
> 4. Connecting radials: I saw mention of one product for attaching the 
> radials, namely the lance johnson radials buss. Certainly, I could mimic 
> this and make my own, but does anyone have another product to suggest? 
> Just trying to compile my options

DX Engineering...do a google.
>
> I am new to this stuff, so please be gentle with me!

No sweat. Have fun. If I have time, I'll model something for ya...but my 
time is not good for the next week...I have to go to the oral surgeon and 
have a molar extracted tomorrow.

73,

...hasan, N0AN