How To Analyze A 49:1 Impedance Transformer (Unun) Using A NanoVNA For An End-Fed Halfwave (EFHW) Antenna Build

Along my journey on understanding radio, electronics, and in particular, the magic of baluns, ununs, and other transformers, I came across this helpful video on how to analyze a 49:1 unun with a NanoVNA. I replicated it with the the “Alleyway Antenna” transformer. This one below…unfortunately I did not take a lot of pictures of the build.

I calibrated the nanoVNA with coax jumpers in place. I set up the NanoNVA like this:

• S11 port to input coax connector of unun
• 3.2 kΩ nominal resistance (3.272 2Ω measured resistance) in series with the unun antenna port and the S21 port of the nanoVNA.
• NOTE: I made a mistake using 3.2 kΩ. This would be roughly 50Ω multiplied by 63 for simulating a 63:1 impedance transformer. I should have used 2.45 kΩ, for simulating 50Ω multiplied by 49 for a 49:1 unun instead. Regardless, I am still presenting the results here.

Below is the datasheet that I constructed for the 49:1 unun under test. The description of the device construction is at the top. It is made in a fairly usual manner, except this time I chose 3 primary windings and 21 secondary, rather than the usual 14 primary and 2 secondary. You can learn more about a 49:1 EFHW transformer build here.

At the bottom are the equations for each of the parameters, and some notes on where to find the data (i.e., the nanoVNA of course!)

One issue that jumps out at me immediately, is the 100.4% transformer efficiency at 7.3MHz. Unless my transformer is truly special, this does not seem accurate.

Below are the results (with closest band markers shown with the triangles) of the analysis graphically using NanoVNASaver software. I am sad to say, it does not appear that this is possible to download anymore and I am glad I retained a copy. You will notice the sweep only goes for 1 MHz to 22 MHz, and has a rather odd bimodal distribution, with the best match showing around the 80M and the 15M bands. And in fact, it is the 17M band marker that is closest to the center of the Smith Chart.

And finally, this is a VSWR to Frequency plot using the 1-port RigExpert 55A Zoom. In this particular case, I have added on some roughly 20 meters of antenna wire in order to test to see how the unun looks with an actual wire radiator attached. I have highlighted in pink which ham radio band that each blue vertical bar represents. You can see “dips” in this trace around the 40M, 20M, and 15M bands, as would be expected for a 40M EFHW antenna. Electically, this radiator wire looks long for its intended use (i.e., the dips in the traces are falling to the left of the desired frequencies, i.e., lower frequencies than desired.) This antenna still needs to be properly hung and tuned in order to obtain the best VSWRs. Regardless, I do expect it will perform well.

This has been a great exercise in antenna analysis for me, even if I should have used a reduced resistance for antenna simulation for a 49:1 unun. Based on my calculations (which in this case I do not actually trust the results), it looks like the transformer has an efficiency across 80M, 40M, 30M, 20M, 17M, and 15M ranging from ~ 82-100%. I look forward to diving into these methodologies more and watching my own understanding of them evolve.

Catch you on the flippity-flip…
KM1NDY