160 meter inverted L radials

My models used bare wire. If using insulated, all dimensions will get about 2% shorter.

It requires an adjustable L-C matching network right at the base of the vertical section to make the antenna usable at various frequencies in the 160m band. Unfortunately, that requires a large variable C and a variable L in a weatherproof enclosure, and you have to run outside to re-tune the network to a new frequency when you QSY. An expensive way is to add a remote automatic tuner like my MFJ-998RT.

If you just prune the antenna so that its lowest Swr is at 1.85MHz, it should be usable without a tuner at either end of the coax at frequencies where the Swr is less than 3. Using a wide-range manual desktop tuner will extend the usable frequency range a bit, but likely not all the way to 2.0Mhz. Don't expect much out of an automatic tuner built-in to a transceiver.

 

I like to use a series coil at the feed point. as the video i posted showed, it is super easy to make a coil and that thing can easily handle legal limit +++. Coil Losses are fairly small because there is only maybe 4 or 5 uH of inductance. Although retuning it requires a trip outside, I get 60-70 khz of total bandwidth at 1.5:1 or less typically the advantage is that an adjustable tap will lower VSWR sufficiently to avoid any power foldback and/or high feedline losses. I can get anywhere in the band at 1.5:1 VSWR or less with that coil - yes, it could go to 1:1 if I added a capacitor but i did not want to introduce mechanical issues oof an air variable or the expense of a vacuum variable -- and best of all, with "just a copper coil" do not have to bother with weatherproof enclosures either. In practice, it can cover the entire 160M band without any risk of moving parts or having to change radiator length with the 5 taps I made on the coil. Literally $9 of copper tubing (ice maker water line) and a ring terminal.... simple simple. works works. good good.

 

Adding a coil only works if the antenna is too short to begin with... It is not a general solution to matching across the 160m band. It takes two adjustable components (L and C) to create a universal matching network.

 

My remote match runs my inverted 160M L for either 160 or 80 meters. On 80 meters an L network is switched in. On 160 meters the antenna is electrically long and another vacuum variable is used to compensate and to raise the antenna radiation resistance.

 

Yes, that is correct. I should have clarified: I purposely make the antenna resonant at the high end of the band *(which makes it shorter in the low end of the band) and by doing so, allows the use of "just a coil" to tune it. As opposed to having to use a much more complex L/C combo that requires water protection not to mention the cost of a good quality air or vacuum variable capacitor. Those vacuum variable caps cost over $100 now days and even a surplus bread slicer is $40+ these days, plus a 2 liter soda bottle to keep it from getting all wet. The coil by comparison is $9.... easily adjusted, requires no enclosure.

It's a balance... the trade off of low cost and mechanical dependability and easy band traversing from cw to phone at legal limit capability with "just a coil" are offset by very negligible loss of base loading ...with a 120 ft wire with a little bit of loading in the cw portion of the band rather than a 130 ft wire. Probably not enough to measure in any real world application.

 

Coil between the vertical feedpoint and radials.

 

I used the model of post #28 and by using some cut and try, I came up with the following. First I found wire lengths that resonate the antenna near the top of the band, actually I chose 1.975MHz. Above that frequency, the feedpoint impedance is inductive, so adding more inductance in-series with the antenna only makes that worse. However, below 1.975MHz, the antenna becomes capacitive, so adding adding inductance as a series-loading-coil just above the feedpoint can cancel the capacitance.

With overall wire length L=172ft, vertical part K=52ft, 4 ea radials length M=30ft, with radials O=8ft agl, and a variable loading coil just above the feedpoint with an inductance Q (uH), here is the required value of Q to resonate the antenna at frequencies below ~1.975MHz:


Since the loading coil alone can only cancel capacitance and it cannot transform impedance like an LC network could, the Swr cannot be reduced below some minimum because even with jX=0, the R term is not 50 Ohms. Here is the lowest Swr50 achievable with just a variable inductor:


I suppose it might have been better to use 2.0MHz as the frequency where the required loading coil becomes zero...

 

I took one more hack at it:

This time I set the "no added inductance" frequency = 2.0MHz, and purposely biased the target feedpoint Z = 60 + j0 at that frequency... This moves the desired ~50 Ohms (to get the lowest Swr50) close to the middle of the band.

The final values are:

wire length L=168.33ft, vertical part K=52ft, 4 ea radials length M=31ft, radials O=8ft agl.

 

yep. Mike. Great model. This proves some of the thinking that I had.

this is generally similar o the results seen with my antenna. Not perfect anywhere but very good compromises

4 x elevated radials 132 ft long, 8 ft AGL (different that your much shorter radials)
1 x elevated radial at 68 ft lnog also 8 ft AGL (for experimenting with the 80M band)
1 x radiator that is 128 ft long, 90 ft go vertical, 38 ft almost horizontal (not my length of radiator is purposly just a little short of 1/4 lambda) at the bottom of the band.
1 x tappable home made series coil with approx 8 uH (resonates at 1810 with 1.2:1 vswr) and reducing inductance via the taps allow for all band coverage.
1 x 3mega ohm resistor from coil to ground for DC bleed-off.

the taps allow no more than 1.2:1 VSWR at pretty much any point from 1800-1900. From there, the coil gets fully shorted (bypassed) and resonates at 1940 khz with min VSWR of 1.3:1 and rises to 1.9:1 at 1.99 mhz All very acceptable considering the trade-offs.

The main goal was to have a very low cost (removed in summer thunderstorm season) 160 solution for CW band but could also go up to the AM window at 1880 Khz and then up to 1920 for occasional SSB nets with limited parts, moving objects, etc

This "works"... for $39 total cost. I have 42 DXCC on CW in my log and only been on 160M for about a month now.

I think the trade off, which I like, is that you have optimized the radiator length AFTER optimizing for a set of very short elevated radials. But as you say, per the model, both methods work and if restriction is on radial length, then this proves one can treat an elevated radial Inverted L sort of like an OCF dipole ...in that the radials and radiator can be of varying length and let the software find the solution.

great work !

 

Joe,

I took your dimensions: wire length L=128ft, vertical part K=90ft, 4 ea radials length M=132ft, radials O=8ft agl, and here are the simulated feedpoint Zs for your dimensions.

Look at the table where Q=0.001 (loading coil shorted out). Notice that the model says your antenna is resonant at ~1.92MHz, but that its FP Z=35Ohms. Making the radials shorter and the wire longer like I did in the previous posts makes R closer to 50 Ohms and results in a net-lower Swr50.

I then asked the simulator what Q would cancel the -jX where it could. Notice that for frequencies above 1.9MHz, there is no coil that helps...You could improve things above 1.9MHz by replacing the coil with a series capacitor (the other part of a full-up LC matching network ).



Some day, when I have more time, I will calculate the required capacitor values. I suppose that it might be useful to reverse the design goals, and come up with lengths where the antenna is resonant at the bottom of the band, and then tuned to resonance only with a series variable capacitor. It is usually much easier to vary a capacitor than vary an inductor.