# Impedance matching transformer

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by W6OGC, Apr 3, 2015.

1. ### W6OGCHam MemberQRZ Page

Instead of a quarter wave line of 75 ohm coax to transform a ~100 ohm loop to 50 ohms, would it be correct to use a toroid or similar transformer with the proper ratio of turns in primary and secondary?

In this case, 100 ohm impedance to 50 ohms would be a turns ratio of the square root of 100/50, or 1.414 or maybe 14 turns to 10?

TIA

2. ### WB5YUZHam MemberQRZ Page

Are you asking this because you want to use the loop on multiple bands?

If so, be aware that the loop has feedpoint impedance of approximately 100 ohms only on the fundamental band. When operated harmonically, the feedpoint impedance is more on the order of 250-350 ohms. So, a 2:1 transformer would, like a 1/4 wave transformer, would still present a correct match on only one band, the fundamental.

This is why so many people use a 4:1 balun to feed their horizontal loop. The theory is that at 300 ohms, a 4:1 gives an SWR50 of only 1.5:1 on all bands but the fundamental. On the fundamental, the SWR50 is approximately 2:1; but since the fundamental is usually 80 or 160m, the operators who choose this method are willing to live with the mismatch, because (a) the coax run is usually short in terms of wavelength on 80 or 160m, keeping losses down, and (b) a horizontal loop isn't very good for weak signal work on the fundamental band, anyway.

Now, on the bands on which the loop is not resonant, that is, those not harmonically related to the fundamental, the SWR50 can be all over the place, and the resultant heating can be pretty stressful on any kind of transformer with a core.

3. ### W6OGCHam MemberQRZ Page

No, different idea. I have been studying delta mono-loops, one loop deltas, and other dangerous ideas.

That's how I understand it, why we use individual ~resonant loops on quads, for example.

I've also spent a lot of time reading and pondering the information on AA5TB's site dealing with EFHWs and how he matches those very high impedance loads to 50 ohms.

A little knowledge is a dangerous thing, it has been well and truly said, so I thought it best to run my ideas past the assemblage of experts here and see if anyone gasped in horror, etc. before I set the weeds on fire here or something.

4. ### W5DXPHam MemberQRZ Page

For a single band, I have used both methods and couldn't tell the difference without measuring the difference.

5. ### KL7AJXML SubscriberQRZ Page

It certainly would work.

The main advantage of "perfect square" transformers, such as 1:1, 4:1, 9:1, etc, is that they're very easy to design and construct. "Odd" ratios are less common.

6. ### KL7AJXML SubscriberQRZ Page

It certainly would work.

The main advantage of "perfect square" transformers, such as 1:1, 4:1, 9:1, etc, is that they're very easy to design and construct. "Odd" ratios are less common.

7. ### KU3XHam MemberQRZ Page

If this helps? I use a 100 ohm 1 to 1 balun backed by a 2 to 1 UnUn. It works perfect on my 30 and 80 meter Delta Loops. I don't use them as multi band Delta Loops.

Barry

8. ### W6OGCHam MemberQRZ Page

I guess that is one way, but aren't you incurring losses in the unun AND balun? As a mostly QRPer, I don't have any db's to squander, so am motivated to figure out the most nearly lossless way to get power from transmitter to antenna.

I don't see any reason to multi band, as with a quad, it involves one more wavelength of wire for each band, and no losses in trying to multi band.

I didn't realize there is such a thing as a 100 ohm balun. Details, please?!?!?!

9. ### G3TXQHam MemberQRZ Page

For single band use your idea is fine. Use a conventional transformer - separate primary and secondary - with the appropriate turns ratio. Use an iron powder core and the losses will be small. Make sure you have enough turns that the winding reactance is at least 5 times the load impedance.

Steve G3TXQ

10. ### W6OGCHam MemberQRZ Page

How about a toroid T-86-2, or so? How does one know that the winding reactance is 5 times the load impedance?