I've read that the commonly used 49:1 (7t:2t) auto transformer for Multi-band End-fed Half-wave (EFHW) antennas has criticisms as not being very efficient. I haven't found any other designs that amateurs use.

I know how to make a 1:1 Common-mode Current (CMC) choke using parallel 100 ohm wire and I understand that one can also make this into a 4:1 transformer with a small modification. I find this balun the easiest to understand.

Is it possible to use 3 of these 4:1 transformers in series to create a "better" 64:1 transformer? I want to use this for 20 watt mobile HF.

The idea being that I could also add a variable capacitor in parallel on the final secondary to tune it to a 49:1.

I guess this is a dumb questions because the 50 ohm characteristic impedance of the 4:1 transformer would force the output to see 50 ohms on the next primary.


1 Answer 1


From memory, I don't have Sevick's book and it's been a long time since I saw it. Maybe someone with the book can correct me.

All high ratio transformers in the book use mutiple cores.

If you want a proper Guanella transformer, one core can only really do 1:1 Balun and 4:1 UnUn, as you say. Everything else is a bit of a hack and results in flux in the core, reduced efficiency etc.

You can (and should) connect several bifilar-wound 1:1 transformers in series to make a high ratio UnUn.
Obviously a 1:1 UnUn doesn't need a choke.
A 4:1 uses one choke and gives you double the voltage. A 9:1 uses two choke for 3 x the voltage.
You can't connect the 4:1s in series, each additional choke gives you one more unit of voltage.

64:1 is 3200 ohms, a very high impedance. What would need such a high feed impedance? I think the transformer design would break down long before this point, the choking impedance of each ferrite and coil will be less than this.

  • $\begingroup$ My goal is to have a 49:1 for EFHW. I found this design and starting to understand it: kn5l.net/kn5lEfhwUnun Thanks for the answer, it's starting to make sense. $\endgroup$
    – wbg
    Jan 25, 2022 at 1:06
  • 2
    $\begingroup$ That's a nice page, it looks like a neat transformer. It's a Ruthroff type, a proper transformer that depends on flux in the core, not done with transmission line tricks. It will have some loss, but it looks reasonable. The only strange thing is the power test done with a 50 ohm load - to measure loss and temperature rise you should ideally connect two identical transformers back to back. $\endgroup$
    – tomnexus
    Jan 25, 2022 at 1:38
  • $\begingroup$ I made one today and had to add a polyvaricap in parallel to the input side. The best I'm getting is 1:1.31 which is OK. Before it was in a box I was able to tune down to 1:1.08 using the cap. This is only for 5 W radio btw. Clearly my build needs improvement. $\endgroup$
    – wbg
    Jan 25, 2022 at 3:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .