I use coaxial transmission line transformers for simple power splitting and combining networks. This is for connecting multiple resonant antennas in array configurations. It's a standard technique that uses a 1/4 wavelength section of coax to match end loads/sources of different impedance. It's become really common in the ham radio designs I'm looking at. Here's a brief note on that: physics behind the impedance matching performed in this case Here is a practical example of one implementation: https://www.qsl.net/dk7zb/PVC-Yagis/2x25_ohm.htm, using 5/4 wavelength cables.

It's so well known and understood (although my theoretical understanding of it is limited), that I wouldn't have a question, except - many designs suggest that odd multiples of 1/4 wavelength should also work, e.g. 3/4, 5/4, 7/4, etc.

I find that intuitively hard to accept for the following reasons. I am somewhat confused and want to know the correct way to implement this. It seems logical to me, that at the end of a 1/4 wavelength transformer, the transformed impedance appears, and so it should be connected only to a coax of the target impedance, not the transformer impedance.

For example, to transform 25 ohm to 100 ohm, I can use 1/4 wavelength of sqrt(25*100) = 50 ohms cable. Call 25 ohms the source, if I want to continue the cable run, I should attach any length of 100 ohms cable to the transformer. That makes perfect sense to me, although I don't fully understand the principle of operation of the transformer itself and just accept that it is true.

But now I'm being told that I could just as well use for example, a 3/4 wavelength of 50 ohms (matching) cable. My conceptual problem is this: I could view that broken down into a series connection of 1/4 wavelength 50 ohms and 1/2 wavelength of 50 ohms. In the first case, there can be a 1/2 wavelength of 100 ohms cable plus any length in addition at the end of the transformer. In the second, there is an actual 1/2 wavelength of 50 ohms cable plus any length of 100 ohms after that.

Questions: How can both cases be valid, or is the theory of odd quarter wavelengths just wrong? If I use lengths like 7/4 wavelengths transformer, what are the consequences? Apparently people are using this without experiencing return losses, but what about signal loss due to the mis-match?

  • \$\begingroup\$ Clearly the longer ones have more ohmic loss so they are not identical. However they may be "close enough". \$\endgroup\$
    – user16324
    Commented Mar 6, 2021 at 19:37

1 Answer 1


A transmission line terminated in a mismatched load of a given impedance will repeat that impedance every λ/2 (electrical length) back from the termination (e.g. here). You can think of your odd multiples of λ/4 transformer as consisting of a single λ/4 impedance transformer (using the calculations you've already described) plus an arbitrary number of λ/2 sections which repeat the mismatched impedance back to the λ/4 section.

You can use this to make a longer matching section. So your example of DK7ZB's stacked Yagi array is using a 5λ/4 transformer to allow larger vertical spacing of the individual Yagi antennas.

There are trade-offs of course. A longer transformer means higher losses. Multiple sections magnify variations in wavelength so a longer transformer will have a narrower bandwidth. You may be able to achieve operation on harmonically related bands - a λ/4 transformer on 2 m might be usable as a 3λ/4 transformer on 70 cm, if you're lucky.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.