EFHW Antenna: A Detailed Analysis (NEC4.2)

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by W9XMT, Mar 29, 2018.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: Subscribe
ad: Left-3
ad: Left-2
ad: L-MFJ
  1. W9XMT

    W9XMT Ham Member QRZ Page

    Below is a detailed analysis of an EFHW antenna with a length of vertical coaxial cable attached at its feedpoint.

    It includes the effects of radiation from the "unchoked" r-f current that could flow along the outer surface of the outer conductor of the coaxial cable.

    Other details about this configuration are shown in the Comment block at the lower left section of the graphic.

    [​IMG]
     
    KD6RF likes this.
  2. AI3V

    AI3V Ham Member QRZ Page

    So you actually modeled a center fed dipole.

    With an extra connection to "ground"

    Rege
     
  3. W9XMT

    W9XMT Ham Member QRZ Page

    However some think that configuration is an EFHW antenna.
     
  4. KD6RF

    KD6RF XML Subscriber QRZ Page

    Thumbs up ;)

    It's interesting to explore the "good lengths" and "sick lengths" (as an old friend used to call it) of feed-line.

    In the case above, common mode current on the feedline is not horrible. In other lengths, the current on the coax shield is higher than on the intended half-wave "radiator" wire.

    Although the common-mode current on the coax shield at the feedpoint is fairly low, it is often incorrectly claimed that the common-mode current is low everywhere on the coax shield - clearly this is not true....

    The coax shield common-mode current is a standing wave just exactly as is the current on the intended "radiator" itself. Which, of course, means that the coax shield also radiates like crazy :eek:
     
    Last edited: Mar 29, 2018
  5. K7TRF

    K7TRF Premium Subscriber QRZ Page

    Rege, I'm pretty sure that's an extreme OCFD with an extra conductor representing the outer side of the coax shield. That Z line in the mid span of the radiator is just the Z reference axis and neither a conductor nor a feed line.
     
    KD6RF likes this.
  6. WA7ARK

    WA7ARK Ham Member QRZ Page

    I do.

    I just got through converting my 40ft vertical standing on a large metal roof into an Inverted-L by adding a slightly downward sloped 32 ft wire. This makes the total electrical length of the antenna 73.25ft.

    efhfInvL.png

    Since I was measuring it with my RigExpert AA-600 analyzer on Ham frequencies anyway, it occurred to me that it would be informative to measure the impedance of this antenna near the frequency where it is one-half-wavelength long, or approximately 6.4MHz. Here is a plot of the readings. The j term goes through zero at 6.405MHz, where R=1420 Ohms. (click to see full image).

    efhwL.png

    I have a toroid auto-transformer designed to feed an end-feed half-wave antenna using 50Ohm coax. I temporarily installed it right at the base of my inverted-L, but under the roof. This transformer has 3 turns on the primary and 21 turns on the secondary, so its impedance transform ratio is (21/3)^2 = 7^2 = 49.

    EfhwTran.png

    Here is a plot of the SWR into the 50Ohm side of the transformer when the top of the secondary is connected to the inverted-L and the common terminal of the auto-transformer (also tied to the shield of the coax) is connected to the metal hangar building.

    efhfts.png

    Here is a plot of R and jX looking into the transformer

    efhwtrjx.png

    Note that this antenna could easily be shortened just a bit to shift it up into the forty meter band. I have something else in mind for it, however.

    So here is a practical example of how to drive a real half-wave antenna using a ferrite RF auto-transformer.

    Mike
     
  7. WA7ARK

    WA7ARK Ham Member QRZ Page

    It appears that the simulated coax shield (the vertical wire in your illustration) is not connected to earth ground at the "rig" end. The simulated common-mode current goes to zero right where it would naturally attach to the rig. This is unrealistic.

    A more realistic simulation would be to create a radial field at that end of the coax shield to simulate all of the ground paths that exist in the typical ham shack. For example, my rig chassis is tied to the green-wire ground conductor in my house wiring, so gets back to the electrical panel ground rod. There is also a entry panel where all my coax shields connect to a separate ground rod. I also have other wiring to other ground rods, e. g. phone line and internet connections.

    After all, most of the controversy about End-Fed antennas centers around common-mode current flow along the coax shield back into the shack, and what that might do to your audio, computer, sound-card, phone line, internet, etc.

    Also instructive, is, if you do get common-mode conducted RF into your shack, where along the coax feed-line between the antenna and the shack should you place common-mode choke(s). Not always obvious, and what works on band-A might create a problem on band-B and vice-versa.
     
    SM0GLD and N0TZU like this.
  8. WA7ARK

    WA7ARK Ham Member QRZ Page

    I take issue with your contention that the common-mode current on the coax feeding an end-fed antenna could ever approach the peak current in the main antenna element. I would like to see you post a simulation where this is the case.

    The amplitude of the standing wave on the coax shield comes from two sources; that which comes out of the Autotransformer (small), and that which is induced into the coax shield by being asymmetrically placed in the near field of the main antenna (likely larger than the other).

    If you use a suitable impedance step-up transformer like the example I just posted above (1:49), the current out of the common terminal of the Auto-transformer (flows onto the coax shield) is small compared to the peak current in the antenna itself.

    It is easy to control common-mode current on the coax shield by using a common-mode-choke (cmc). In fact you might want more than one; one cmc to prevent distorting the radiation pattern by blocking current on the vertical segment as shown by the Thread Starter (or to prevent it from picking up local noise), the second cmc to prevent common-mode current from flowing into your shack and screwing with other equipment.

    Mike
     
    Last edited: Mar 30, 2018
  9. W9XMT

    W9XMT Ham Member QRZ Page

    As stated in the Comment section of my graphic and shown in the wire model at the right side of the graphic, the bottom of the coax shield is connected to the top of a buried, 8 ft vertical ground rod. The r-f current of that path drops to zero at the bottom of the ground rod, not at the bottom of the shield.
    I used NEC4.2 for this analysis, and it calculates the losses that exist for the conditions stated in the NEC model (for the frequency, earth conductivity, and buried wire configuration).
     
  10. WA7ARK

    WA7ARK Ham Member QRZ Page

    So what is the magnitude of the current right at the top of the ground rod in Amps?

    What is the peak current along the horizontal wire?

    Should be easy to extract those two numbers from your sim.

    Taking this one step further, if you have the time, add a common-mode-choke with a series impedance of 1000+j1000 just above the ground rod, and re-sim. What is the peak current into the ground rod after the choke is added?

    Mike
     
    Last edited: Mar 30, 2018

Share This Page