Calculations of end-fed vertical performance and questions about unun/losses/antenna length

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KW4TI, Jul 24, 2016.

ad: L-HROutlet
ad: l-rl
ad: L-Geochron
ad: Left-3
ad: abrind-2
ad: L-MFJ
ad: Left-2
  1. KW4TI

    KW4TI Ham Member QRZ Page


    I am a new ham and an electrical engineer, and I am trying to put up a vertical antenna along a very tall tree, about 80 feet tall (a Loblolly pine). I have buried 14 radials about 30-40 feet long, and I am hoping to get at least 40-10 m bands, with 80 m being a stretch. I have read about the 33ft/43ft verticals and using a 4:1/9:1 unun, so that is what I'm trying. I plan on starting using an antenna tuner in the shack (MFJ-929) and perhaps moving the tuner to the base of the tree in a weatherproof box if there is problems. My coax run to the bottom of the tree is about 150 ft of LMR 400 to try to minimize losses when antenna matching in the shack. I have some restrictions from the XYL and potentially neighbors about the antenna, so I am trying to keep it discrete (no yagi farm suggestions please), and keep the coax out of sight. My antenna wire is a 26 AWG magnet wire painted black hanging from woven 50 lb test spectra fishing line from the tree, so it is very hard to see from a distance.

    I have a number of questions. The first is what material I should be using from the unun. I have wound a 9:1 trifilar Ruthroff unun with taps for 4:1 and 9:1 on a FT240-43 core. I chose this over a FT240-61 core because I thought that the higher choke impedance was necessary on the lower bands more than the likely losses on 15 and 10 m. I decided against the instructions I saw on the net for using a T200-2 or T200-6 core because from what I understand, the choke impedance of these are very low and therefore not effective at 80 m. One of the issues I read with using a lot of winds on the unun is that the capacitance between the windings can cause a resonance if too many are used, and I thought that with FT240-61 too many winding would be needed to get enough choke impedance at the lower bands. I placed a 450 ohm resistor on my FT240-43 and measure very close to 1:1 SWR for 80-15 m bands, and 1.5:1 for 10 m band, though it could be due to the resistor reactance (I could only measure the reactance at 100 kHz using my LCR meter).

    I also wanted to figure out what wire length to use for the antenna. I don't have EZNEC and don't want to have to model every single wire length, so I use the idealized formula the feed impedance of a dipole from section 8.5 for Constantine Balanis's antenna theory book, and halved the impedance due to the ground plane of the vertical. Then I added the estimated resistance of the actual ground to the feed impedance to account for ground losses (I guesstimated about 20 ohms ground resistance, because I live in a place with poor soil resistivity). I used the 26 AWG wire diameter to calculate the reactance of the antenna (the resistance doesn't depend much on the wire diameter). I also calculated the performance for 4:1 (200 ohm) and 9:1 (450 ohm) idealized unun output impedance.

    So the first thing I did was to calculate the antenna lengths (in meters) that are not near a half wavelength for the 80 to 10 m amateur bands (160 m is hopeless I'm sure for the wire lengths I'm considering):

    2.9000 3.0000 3.1000 3.2000 3.9000 4.0000 4.1000 4.2000 4.3000 4.4000
    4.5000 4.6000 4.7000 4.8000 5.8000 5.9000 6.0000 6.1000 6.2000 6.3000
    6.4000 6.5000 6.6000 6.7000 7.4000 8.3000 8.4000 8.5000 8.6000 8.7000
    8.8000 8.9000 9.0000 9.1000 9.2000 9.3000 9.4000 11.5000 11.6000 11.7000
    11.8000 11.9000 12.0000 12.1000 12.2000 12.3000 12.4000 13.6000 13.7000 14.6000
    14.7000 14.8000 14.9000 16.5000 16.6000 16.7000 16.8000 16.9000 17.0000 17.1000

    So I found that indeed there is a wire length at 13 m that corresponds to the 43 ft vertical. I also found a slightly longer length at 17 m (56 ft) that might do a better job at 80 m and still fit on the tree.

    I then calculated the feed impedance including ground losses using the textbook formula, and generated these plots of the impedance, antenna efficiency with 20 ohm ground, an the VSWR from 1.8 to 30 MHz.

    As you can see, 17 m does look promising. However, from what I understand the antenna length may need to be tweaked a little bit because my antenna is not ideal. The 9:1 transformer seems to be more tolerant than the 4:1 transformer.

    So another question I have is, basically, how useful the analysis of eliminating the wire lengths that are half a wave length. It seems to somewhat confirm what I have already read, but perhaps there are some factors I'm missing here.

    Another issue is should I use a common-mode choke at my tuner? I have a FT240-31 core that I would like to turn into a common-mode choke by wrapping 10 turns of RG313 coax around it, hopefully to reduce the RF feeding back into the radio, as well as perhaps reduce noise picked up on the coax.

    Finally, would a 4:1 or 9:1 Guanella balun be useful here rather than a Ruthroff unun? I am feeding an unbalanced antenna with an unbalanced coax, but perhaps the lower loss of the Guanella balun might buy me a little?

    Sorry for the extended post but I thought the answers to some of these questions might help others as well who find themselves trying to hang a tall wire from a tree (especially here in the North Carolina where they can be found). I included the MATLAB code I used to make these plots to help anyone interested in this.



    Attached Files:

  2. KA0HCP

    KA0HCP XML Subscriber QRZ Page

    Your thoroughness as an engineer in seeking the optimum design is admirable!

    From the technicians practical perspective:
    -Ham axiom: "Anything that will conduct electricity can be used as antenna"
    -Ham axiom: "If it loads up you can make contacts"
    -No, use on 160m is not ruled out. You will just have low efficiency.
    -There is no such thing as an ideal multiband antenna
    -Nearly every ham must make trade offs and concessions to his particular situation

    -Propagation is the great equalizer among antennas.

    Hams have successfully used beer cans soldered into vertical elements, bed springs, window screens and lawn chairs for making contacts. I have worked CW with a WWII aircraft transmitter in my townhouse basement with a lightbulb dummy as an antenna. :)

    Finally: Don't drive yourself crazy seeking perfection!

    Antenna experimenting can be a lifelong interest. Have fun. b.
    K7TRF likes this.
  3. K7TRF

    K7TRF Premium Subscriber QRZ Page

    Yeah, a lot wrapped up in your post. The advice above is good, especially for your first antenna. Sure get a good handle on what you want to build but don't over analyze nor over optimize just to get on the air. There's always time to play with alternative antenna designs and to compare and frankly without a baseline for comparison it's really hard to know if the theory plays out. IOW when you have only one antenna it could be awesome or awful but unless you have a reference antenna for comparison it's hard to know.

    In terms of your questions:
    - You can download a free version of EZNEC that's full featured in terms of analysis but limited in terms of the number of wire segments you can model. For an antenna like what you're considering or simple dipoles, doublets, verticals and such the free version is fine. For someone that is clearly interested in analysis it's worth the very easy learning curve to figure out how to use EZNEC and it may open up a lot of good antenna possibilities:

    - In almost all situations but especially for antennas that are not very well balanced I'd prefer a current mode balun to a voltage mode balun. So yes I'd recommend a 3 core Guanella as opposed to a Ruthroff for your antenna. You could as you suggest chain the two with a 9:1 voltage balun in series with a 1:1 current choke. If you go that route I'd place them together at the antenna feed point instead of one at the shack and one at the feed point. It's best to minimize common mode currents flowing on the feed line (assuming you have a good RF Return/counterpoise at the feed point) and choking common mode at the feed point should also minimize common mode in the shack itself. So I'd either go with a 3 core 9:1 Guanella or place a Ruthroff 9:1 at the feed point and a 1:1 current choke directly between it and the feed line to the shack. Depending on how you manage your winding impedance (e.g. coax vs bifilar) you could place the two baluns in either order but it's probably simpler to build the 1:1 current choke out of 50 ohm coax and place it on the feed line side even if it's just the feed line itself wrapped through a large core or a clamp on core.

    - Yes, it's hard to build a really broadband balun as too few windings and you lack choking impedance on the lower bands, too many windings and you risk excess parasitic capacitance and resulting lack of choking on the highest bands. A couple of octaves of good choking performance isn't too hard but it's hard to say build a good balun that spans 160m to 10m without some compromise on one end or the other. This chart is pretty handy for figuring out tradeoffs in current mode balun design: with the goal being sufficient choking impedance on the bands of interest and ideally predominantly resistive as opposted to reactive choking (look for the frequency ranges with black underlines to see where various chokes are mostly resistive).

    - Remember the real goal of feeding an end fed antenna with a 9:1 balun isn't really achieving a perfect match on all bands, it's limiting the range of expected impedances across a wide range of bands so that the rig or shack tuner has a fighting chance at loading into the system and feed line SWR is reasonably low hence feed line related losses are reasonably low. IOW, you may get lucky and see good SWRs (e.g. below 1.5:1) across all bands with a good 9:1 balun, a decent system of radials or RF Return (i.e. counterpoise) and good choices of wire lengths, but it's just as likely you'll see a bit higher SWR on some bands and that's fine as long as your rig can load into it which depends a lot on the rig. For instance vintage tube rigs with PI network output stages can load into a fairly large range of SWRs, modern rigs with internal auto-tuners can also load into a fairly large range of SWRs but modern solid state rigs without auto tuners typically fold back output power if the SWR exceeds about 2:1 and may need help with an external tuner if you don't get lucky with antenna system impedance on all bands. Also remember that calculated antenna wire lengths ignore environmental effects like capacitive loading due to dielectric or conductive objects near the antenna (houses and other structures, trees, large conductive objects, height above ground, etc.) so don't expect calculated wire lengths to be right on the money.

    The antenna you describe will almost certainly work but don't be surprised if it takes a bit of field tweaking to get it working as expected. That gets back to the post above, at some point it pays to just build the antenna, see how it behaves and tune/adjust it as necessary.

    Good luck,
    KA0HCP likes this.
  4. WB5YUZ

    WB5YUZ Ham Member QRZ Page

    Welcome to the hobby. Thanks to your education you will enjoy a tremendous advantage over us lay hams in actually understanding things. I do have a few observations, however:

    (1) If you haven't done it yet, get some antenna modeling software and check the patterns of the various lengths of wire you are considering, on all bands. Since you are making a vertical that will cover the longest wave bands, pay particular attention to the shortest-wave band patterns. You will probably find that a nice, long vertical which makes an efficient radiator on 80 or 160m is going to have an unprofitably high-angle radiation pattern on 15-10m. Remember, though, that those higher-frequency bands aren't going to be open that much in the next few years, giving you time think about a different antenna to cover those bands. You could also use traps, or parallel/close coupled feed of multiple vertical elements.

    (Below: the radiation pattern of a 50' elevated vertical on 10m, superimposed with the pattern of a simple dipole a half-wave high.)


    (2) Usually, any choke intended to reduce common-mode current in the outside of a coax feed line is installed at the feedpoint of the antenna, not the shack end of the run. The shield at the shack end of the run is supposed to be grounded, anyway, as per NEC.

    (3) Generally speaking, an elevated vertical will outperform a ground mounted one, unless the ground mounted one is in the clear, off by itself. If you live in a crowded suburban or urban environment, and you have the space/means to do so, consider elevating the feedpoint and using tuned radials for each band of interest. In addition to getting the maximum current point of the antenna above ground clutter, it also can increase efficiency by lowering ground losses.
    KA0HCP likes this.
  5. W6OGC

    W6OGC Ham Member QRZ Page

    I'd suggest instead of the MFJ 929 a remote autotuner at the feed point which will get you a low(er) coax loss on that 150' run. That will give you fewer worries about low SWR. There are several choices depending on how much power you plan to run.

    Some favor SCG units, some like LDGs. I used an Icom AH-4 that matched everything I connected it to. It is limited to 100 watts or so. You might look into maximum length of the control wire, voltage drop, etc. if that might be a factor. Mine was always fairly close and not a problem.
  6. KD6RF

    KD6RF Premium Subscriber QRZ Page

    As others have said already - good job on first cut analysis!

    Lots of good advice above - I would add only a few items -

    One is a slight conditional disagreement with my friend Dave "the other RF" :) and I wouldn't worry too much about common mode currents meaning that a simple unun is just fine, *unless* you have a very short coax run and/or you don't plan on having a good radial field (which would be bad!). I run unun's on my Inverted-L's and there is no measurable shield current at the shack entrance. I do use decent radial fields (at least 15 radials).

    Another is to not worry too much about higher orders of 1/2 resonance - really only the first one or two are the really large high-Z bumps that needs to be avoided - this means you don't have to worry too much about parking the 1/4 wave length in the 4.5 or 6 MHz regions.

    Finally, consider going inverted-L - a 45 x 45 for 160M - 10 M, or a 23 x 22 for 80 M - 6 M plays well on the higher bands, giving 4 or 5 dBi or so nice low angle horizontally polarized radiation on those higher bands from the upper section. Better than just a vertical on those mid and upper bands, but not quite as good as a dipole at the same height.
    Last edited: Jul 24, 2016
  7. KW4TI

    KW4TI Ham Member QRZ Page

    Thanks for your replies and your helpful advice.

    So I have a few more questions,

    1. I was considering buying an outdoor autotuner, but decided to start with a tuner in the shack. I considered that either I could route all of the control lines to the tree, or inject the power using a bias tee over the coax. If there are no control lines, then how do you know when you key up your radio with full power that the autotuner has matched the antenna? For example with the MFJ, you have to set up a learning mode where you send low power to the autotuner for various frequencies and it learns the antenna, and then full power could be used. But what if it rains and the radial system or wire has a different impedance, or the wire moves a bit? How much does the SWR change, and how often should one "relearn" the frequencies. I suppose one could watch the SWR meter on the radio to see if it changes greatly. The MFJ 929 has a bias tee built in, and I was wondering if I could put that into in a weatherproof ABS junction box. From what I have read, most of the MFJ outdoor tuners and CG-3000 don't have great weatherproofing or build quality, and only the SG-230 seems to be favored because it is designed for marine use, but it is rather expensive ($500). I am not sure if the MFJ 929 can tolerate extreme heat or cold if in an ABS box, but I think it should be possible to put it in a better enclosure than most of the outdoor autotuners are sold with. Does this seem like a reasonable approach? The alternative is using the tuner in the shack, which I calculated probably incurs a 1.5 to 2 dB loss at 5:1 SWR at 20 m.

    2. Elevating the vertical antenna seems like a good idea, but one thing I have not figured out is how to do it without having unsightly cables or boxes hanging in the sky. Right now my wire is extremely thin and unnoticeable. For example, if I made a vertical dipole, I could bring twin-lead in from the house to the tree, but then this big piece of twin lead would be hanging in the sky. Would it be possible to make a nearly invisible piece of "twin lead" using two thin magnet wires glued to pieces of transparent plastic to hold them at a fixed distance at regular intervals? If I hang my unun or autotuner from the tree at 20 or 30 feet high with the coax, that will be noticeable. I could use a really thin coax like RG316 and that might be an option, and then tie the shield of the coax to the ground radials at the bottom of the tree. But a thicker suspension string would also be required to support the increased weight of the coax and unun as well. Are there any good solutions for hiding the feed? Are there any solutions like running a second wire along the radiating element to prevent

    3. I also considered an inverted vee, using the tree as the support, but I have to get the feed up the tree again without it being noticed, and then bring wires down at an angle of 45-60 to the ground from the tree to places where landscapers or birds won't crash into the nearly invisible wire, or the neighbors will notice. It's like the old joke about hams: you'll have friends all over the world, just not in your neighborhood.

    4. I might try the Guanella balun to see if the frequency response is a bit flatter as well as prevents common-mode currents, and its not too hard to make, and perhaps it will buy me something.

    I know I am probably overthinking this, and I've already started to make a few contacts on 40 m, but from what I've heard the band conditions are not great lately, but I am enjoying learning about all of this, and like to understand it.


  8. K7TRF

    K7TRF Premium Subscriber QRZ Page

    All your ideas above can work. Folks have definitely remoted indoor tuners in suitable enclosures, similarly you can use stealthy antenna wire like enamaled or even bare 18 to 22 AWG with the only real downside being that thin wires snap easily. Similarly you could definitely homebrew stealthy balanced feed line as there's nothing really special about balanced feedline beyond maintaining a consistent spacing. Sure the exact feed line impedance will vary with things like wire diameter, insulation or lack thereof and spacing but in the case of a remote tuned Doublet the specific feed line impedance doesn't really matter.

    The biggest challenge with stealthy balanced feed lines is that you really want to keep the line away from objects, especially conductive objects but even running it right along the bark of a tree will likely impact the line's performance. Coax doesn't suffer from that kind of proximity effect so in some ways coax is better for stealthy installs as you do things like staple it to your house or run it right up a tree trunk without worries.

    In terms of remote tuner performance when you use a remote tuner without control line feed back (like some of the mobile/marine tuners have) you just watch the SWR meter on your rig and when it settles down at a reasonable SWR you're good to go. You don't need to see 1:1 on the SWR meter, just low enough to keep the rig happy which is generally below 2:1 though it's nice when an antenna is at or below 1.5:1 on frequencies of interest. So if you remote a tuner running off a bias-T power feed you just key up at lowered power, watch the SWR meter if it starts dropping in steps, the tuner is tuning. When it settles at a reasonable SWR and stops changing (usually takes only a few seconds or less) then you can go to full power and start transmitting.

    If you run an inverted-V, try to tie off the ends at least 8 to 10 feet above the ground or higher to keep them out of harms way. So if you use the tree as a center support it's good if the ends can attach up in the branches of another tree or to the side of the house or garage or similar. Remember the highest voltages will be at the ends of an antenna like a dipole or inverted-V so insulate the ends and keep them at least a few inches if not a foot or more from anything that could conduct and away from where people or pets could touch them while you transmit.

    Your original idea of a wire vertical with a field of radials is a good idea. In that case you could remote tune it with a tuner concealed in something like a hollow landscaping rock like they use to cover sprinkler controls and just run a light gauge stealthy wire straight up to an insulator and then some cord to a tree branch. Something like that could certainly work.

    Good luck and have fun experimenting,
  9. KA0GKT

    KA0GKT Ham Member QRZ Page

    The August, 2016 issue of QST has an interesting antenna designed to be hung in a tree.
  10. N7EKU

    N7EKU Ham Member QRZ Page


    Personally I would say you are diving in to deep waters on this first attempt. It can be endlessly complicated setting up a compromize length vertical with an adjustable tap balun on a long coax run to a tuner so you can use it on every band available to you. Even the sentence is complicated! Trying to figure out the radiation pattern on the X/X wavelength vertical that will result depending on the band you are on would also be a real challenge.

    You might start with a 40M 1/4 vertical. This can be adjusted to give a fairly low SWR at the feedpoint for using it on both 40M and 15M. You could also set up 1/4 verticals set around the sides of the tree and run them together a la a fan dipole. You could also just use a remote coax switch at the feedpoint to these so you could switch between your verticals as desired. With such skinny wire, I doubt anyone would notice at all.

    You will be missing out on 80M! With such a tall tree, I would certainly not want to miss out on having a real 80M 1/4 wave vertical on that band. Not many people can do that!

    Best of luck and have a lot of fun!


Share This Page