All of these concerns are just flies on the back of the cow if the
ground/radial/counterpoise system is not efficient. OTHERWISE...
If you model a lossy ground/radial/counterpoise into the equation and make
those adjustments, the ONLY thing that makes things better (other than
fixing the ground/radial/counterpoise) is TO LENGTHEN THE WIRE. That is
because a bad ground/radial/counterpoise with a series resistance of say 25
ohms will require the feed R to go up for more of the power is dissipated in
the wire instead of the ground. This change will swamp all other
Guarantee your ground/radial/counterpoise series R is down to 2 or 3 or 4
ohms. Efficient ground/radial/counterpoise is the overwhelming good
performance bad performance indicator for a 160 antenna. NOTHING in second
place. Nothing in third place. Nothing in fourth place. Do that FIRST.
Do it WELL. THEN worry about the other stuff. That said....
I'm putting up a 160 L that will be up 90' and out 110 feet. Counterpoise
is at 8 feet above ground. 90' is determined by that particular tree as is
the 110' by another particular tree. Note that while models can be
notorious for missing absolute values, they are very good for showing trends
and comparisons as elements are varied. I have varied the length of the
horizontal section and here are the model numbers at 15 degree takeoff, plus
other considerations. (EZNEC Pro/4 5.0.39 running NEC4 core, real high
length gain 15 deg est feed R Current max gain all directions
horiz tune out X ft above feed 360x90 deg
3/8 110 -.06 dB 90.2 +60 max
105 -.02 81.1 +54 -.13
100 -.01 73.6 +49 -.27
>>>> 95 max 67.3 +44 -.41
>>>> 90 max 61.9 +38 -.54
85 -.02 57.3 +33 -.67
5/16 80 -.04 53.3 +27 -.78
75 -.08 49.8 +19 -.89
70 -.12 46.6 +8 -.99
65 -.17 43.8 bottom -1.09
60 -.20 41.2 bottom -1.18
55 -.26 38.9 bottom -1.26
1/4 48 -.36 35.9 bottom -1.38
In the last column, note that while the low angle radiation only changes
0.36 dB over the entire range, the gain to ALL directions and angles
improves a much larger 1.38 dB going toward the longer horizontal. In this
case, using all the length saves me 1/2 a dB in all directions overall while
only loosing 0.06 dB at 15 degrees. That's an easy pick for me because I
also work things besides far DX.
That the best radiation occurs with the current max at the bottom is an
urban legend. A lot of important phenomena are seen when you run the
models. As in some other number of instances, the maximum low angle gain
for an inverted L occurs when the current max is a little over halfway up
the vertical stretch.
2010/9/23 mrtman777 <email@example.com>
> Scott KB0FHP writes:
> What dimensions are best - 5/16 or 1/4 wave? Both are indicated in the
> What is simpler to match?
> George K8GG writes:
> The easiest method to get 50 ohms is to make a 1/4 wl Inverted-L,
> and ignore the SWR at the feedpoint, feed it directly with good quality 50
> ohm coaxial cable into the shack and then use an L-network antenna tuner
> to match the impedance at the shack end of the coaxial cable to 50
> ohms. TenTec model 229 or 238
> antenna tuners, or some of the military surplus L-network antenna tuners
> work very well for this application.
> There will be very little loss from SWR in the coaxial cable at 1.8
> George, I agree with you completely. First, let me add a supporting
> for the 1/4 wl inverted L.
> Since the inverted-L will require a matching network whether it is 5/16 or
> wl, that leaves us using other factors to decide which length to use. To
> increase the DX performance of the antenna, you want to maximize the low
> radiation. We know that the current in an antenna element is what causes
> radiation, not the voltage. With the 1/4 wl antenna, you get the maximum
> current at the feedpoint. Obviously, and as is indicated in the
> this typical 1/4 wl inverted-L will have alot more radiation coming from
> vertical portion than coming from the horizontal portion, as the highest
> percentage of current is located there. By increasing the length of the
> inverted L to 5/16 wl, you move the current maxima up the antenna. If you
> lengthening the antenna, you would eventually get the current maxima to
> the junction of the vertical and horizontal parts. In this case the
> horizontal part would have as much current and radiate as much as the
> part. So by lengthening the antenna over 1/4 wl, you decrease the
> percentage of
> the total antenna current that is flowing in the vertical portion from the
> wl case. This isn't what we want. We want maximum current flowing in
> the vertical portion for improved DX performance. By using a 1/4 wave
> inverted-L over a 5/16 wave inverted-L, you gain some low angle radiation
> that hard DX path, at the expense of a bit of high angle radiation for
> stateside, where the signals are always way stronger. This few extra
> dB can
> serve you much better in working DX than for working stateside. My 1/4
> still does fine stateside, and performed well in ARRL 160 last winter.
> Second, my reasoning for matching at the shack as George suggests instead
> matching at the antenna:
> Say for a long 1000ft. run of RG-213, you will only lose an additional .35
> for a 2:1 SWR at 1.8 Mhz. I don't imagine that your run will be nearly
> long. This kind of loss is neglible.
> You don't have to bother weather proofing your matching network.
> After laying radials all day, when you are ready to tune up the antenna and
> it out, sitting there in your ham shack tuning the network sure beats being
> at the antenna feedpoint on the ground doing it. (Don't ask me how I know
> this... hihi)
> If you have problems this winter with the network, do you want to be out in
> cold troubleshooting it, or in your warm shack sitting down with a cup of
> at your workbench?
> I believe all this is worth losing an extra fraction of a dB in the coax.
> The attached picture shows the network I use. Its values were calculated
> ARRL's TLW (Transmission Line for Windows) that came with the Antenna Book,
> and the coax length was taken into account. I determined the exact values
> a variable capacitor and a clip lead jumper on the inductor and replaced
> with set-value components to save space. All of these parts are out of my
> junkbox, and it was built in an afternoon. Ignore the relay. It is simply
> bypass switch for the network, so I can take it out when using my L on
> bands (it is trapped). Note: The inductor stock is homebrew patterned
> after a
> recent article in QST by Barry Shackleford, W6YE.
> 73 and GL,
> Trey KJ4FDV
> UR RST IS ... ... ..9 QSB QSB - hw? BK
UR RST IS ... ... ..9 QSB QSB - hw? BK