ad: Alphaant-1

Why 49:1 or 64:1 for EFHW UnUn

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by WK7L, Jan 23, 2021.

ad: L-HROutlet
ad: l-rl
ad: HRDLLC-2
ad: Left-2
ad: MessiPaoloni-1
ad: abrind-2
ad: Left-3
ad: L-MFJ
  1. WK7L

    WK7L Ham Member QRZ Page

    I'm starting out with a 64:1 precisely because the generally quoted range of 3000-5000 ohms for EFHW antennas gets closer to 50 ohms with a 64:1 than a 49:1. It won't be perfect, but I want to be as "good enough" as I can
  2. WA7ARK

    WA7ARK Ham Member QRZ Page

    If you think of this antenna as an "off-center-fed dipole" instead of an "end-fed-wire" you get to choose any feedpoint impedance you want from <100 Ohms up to about 4000 Ohms. Only the lengths of the wire to the right and left of the transformer change to accomplish that.

    Personally, I think that choosing a feedpoint impedance of ~2000 Ohms is much easier to manage than striving for a higher impedance... That requires a transformer with three primary turns and S=16 like I showed in post #6. To make a swr50=1.003 at 7.15MHz, using uninsulated wire, that requires a radiating wire of 53.19ft and putting a CM choke 4.3ft down the coax from the transformer...
  3. WK7L

    WK7L Ham Member QRZ Page

    Although I had not initially thought about using different winding ratios to achieve different SWR matches, the page at seems to address this idea quite nicely. The author
    suggests that you can tap the last few windings on a 3:24 toroid and switch between them to get a better match. He does this on what appears to be a current transformer, though I don't see any reason that it wouldn't work with an autotransformer too. Changing the impedance match should, in theory, change the SWR. I could see how this might even be leveraged for a multi-band 80-10 EFHW antenna that covers both the CW and phone portions of the 80 meter band with just the flip of a switch. I'll have to try that out when I get some time.
    KB8CR likes this.
  4. KG2RG

    KG2RG Ham Member QRZ Page

    all great explanations, I'll add one more.

    49:1 transformers transform 2450 ohms down to 50 ohms. the winding is 2 primary turns and a total of 14 turns. that's 14:2 and a ratio of 7:1. Sq the 7=49
    50 ohms x 49= 2450 ohms. if the impedance at the feedpoint of the half wave end fed is exactly 2450 ohms, then your 49:1 transformer will transform that down to 50 ohms..
    but its never 2450 ohms, sometimes its more or less. location, materials, half waves at other bands, instal, configuration, all contributing factors to what the impedance will be at the feed point once installed.

    you could try a 49:1 or a 64:1 and see what the match is after install. if your showing 80 ohms, your not transforming enough, ad a turn on the winding. if your impedance is 30 ohms, your transforming to much, pull a turn and check it again, tune it up. if you can get 50 ohms, 1:1 sir and 0 reactance on a frequency in the middle of the band, your transformer will have some bandwidth range. as you move up and down the band from where your match is perfect, you will notice your SWR, reactance and ohms begin to change and get less desirable. but it still may be under a 2:1 swr and fine to operate with some level of efficiency. but if your match is perfect at the top or bottom of a band, you will be using your tuner at some point.

    one other thing about toroids: FT240-34 is pretty efficient, while the 140-43 is less efficient and a 64:1 ratio works better on them then does the 49:1
  5. WD8DSB

    WD8DSB Ham Member QRZ Page

    Interesting topic, and I would like to add another twist. Looking at some data from W8JI, and also doing some 4NEC2 modeling of my own, the feedpoint impedance at resonance is highest when the EFHW antenna is operated on its fundamental band (example: when using a 40 meter EFHW antenna on 40 meters versus using it on its harmonic bands). Because of this a 64:1 transformer looks to be a good pick if operating the EFHW on its fundamental band where its feedpoint impedance is highest, but if operating the EFHW on its harmonic bands where the impedance drops then a 49:1 transformer might be a better choice. I personally use a 64:1 transformer on my 40 meter EFHW antenna that I use primarily on 40 meters and this works really well. At times I've also considered using a 56.25:1 ratio as a compromise (example: 2 turns on the primary and 15 turns on the secondary of my home brew transformer). There are lots of variables that come into play, and no one correct choice as others have mentioned. Just FYI and 73, Don (wd8dsb)
    WA7ARK likes this.
  6. WB2UAQ

    WB2UAQ Ham Member QRZ Page

    This is an interesting topic however, my EFHW 66 to 67 feet long, tested at three locations and at various angles WRT horizontal, looks like 3700 ohms in parallel with -j2675 ohms at 7 MHz (8.5 pF shunted across 3700 ohms). Zs = 1270-j1757 where |Zs| = ~2168 ohms.

    Yes, the equivalent parallel resistance is within the range mentioned above but there is a significant reactance here as well. I have raised this many times before but no one has commented or offered an explanation. How can we just base the turns ratio (or the turns ratio squared) on simply a resistance? (the shunt C of 100-150 pF is needed because the primary inductance is too low) Sorry if I was asleep at the wheel and someone has:)

    4NEC2 offers a higher shunt C (~16 pF) but sort of in the ball park (I have to remember to move the ground conductivity around to see what its impact is).

    Has anyone else measured the impedance looking into the EFHW wire or back into the 1:49 or 1:64 transformers (with a 50 ohm resistance applied across the input)?

  7. WA7ARK

    WA7ARK Ham Member QRZ Page

    That just means you never tuned it to resonance, or found the actual resonant frequency by sweeping it.

    Here is a free space model of a near-end fed wire. There are two variable lengths; L (the short wire, eventually to be replaced by CM current on the outside of the feeding coax) and M (the long wire).

    I am designing for 7.15MHz. There is some length L and M that makes the wire resonant (jX~=0) at the same time that the R at the feedpoint is 2450 Ohms (or whatever R you want, within reason). I use the optimizer in AutoEZ to find L and M (in ft):

    Note that I show the Swr relative to 2450 Ohms, my design goal. If connected to an ideal 50:2450 Ohm transformer, that combination of L and M would have a Swr50 of 1.00. Also notice that the peak gain of the antenna is 2.11dBi, which just happens to be the same as a free-space, resonant, half-wave dipole, which is exactly what it is! Here is the current magnitude along the wire. Look familiar?

    Now that we have a "resonant" end-fed, I show what happens if I keep L fixed at 7.35ft and vary M:
    This shows the sensitivity of R and jX to changing M. This shows that M can be changed to bring the system to resonance (make jX small).

    There is another whole discussion that we can have about making L act as though it is about 7.35ft long if it is replaced by CM current flowing on an arbitrarily long piece of coax.
    Last edited: Jul 26, 2021
    N4OKN, WB2UAQ and W9IQ like this.
  8. K8JD

    K8JD Ham Member QRZ Page

    Tuning For Compromise
    Even that brilliant math can not predict how you have your antenna installed at YOUR QTH! It is all a compromise.
    IDEALLY you need a tuner like a PI net up where the coax hits the wire to adjust, to match the wide variations of Z and X you get on multiple bands with a fixed length of antenna wire.
  9. WA7ARK

    WA7ARK Ham Member QRZ Page

    ...and you know this how? I have never seen you post anything but pure conjecture.
  10. K8JD

    K8JD Ham Member QRZ Page

    60+ years of building and observing antennas and reading the antena theory books ?

Share This Page