Testing a "balun" with Analyser

[VK2CSW]

Hi All,

I am almost scared to ask this.

I was having a fiddle around with some ferrite toroids of unknown heritage - I was winding some speaker wire around them in a so-called 1:1 or current balun configuration. (I am really not interested in the likely barrage of posts pointing out that they are/are not actually baluns, etc)

I was simply playing around to see what differences changing the number of turns and spacing would make.

To measure it all I was using a VK5JST analyser.

Now to simulate the imaginary antenna I placed two 100 ohms resistors in parallel across the antenna side of the "balun".

I was getting all sorts of weird and unexpected values on the analyser for impedance and reactance. Now if I placed the resistors directly across the analyser it all looks ok - 50 ohms imp and 0 ohms (ish) reactance and vswr 1.01:1.

Is using the resistors as a dummy load a valid test?

[VK3ALB]

To answer your question,

Yes, you can use resistors on the far end of a balun and get meaningful results from the analyzer.

BUT

To understand your observation you need to dig a bit deeper.

How is it that you are getting erratic readings? You've already proved the resistors are fine to - say 30MHz so what gives?

Is it really a transmission line you've wrapped around the core?
Is the core material correct for your application?

They look so simple but there is a lot to learn about baluns.

[VK5ZLR]

toroids of unknown heritage -that could be half the problem.

[VK2AVR]

Yep, perfectly valid to test that way. I verified my 4:1 guanella balun using a pair of 100 ohm 1/4 watt resistors in parallel. The SWR curve was 1:1 at 7MHz gradually rising to I think 1.3:1 at 28MHz. Which was pretty much what I was expecting given the core choice (mix 43)

Wirewound resistors will ruin your day because they are inductive. You need resistors that are non-inductive at the frequency you're using.

Ferrites of unknown heritage really need to be characterised by measuring them directly (not trying to build a balun straight off the bat). They could be VHF mix for all you know!

[VK2CSW]

Thanks for the replies.

To further explain:

I was just pottering in the shed on a lazy Sunday - I need to build a balun / choke for a 20m dipole that is to replace the G5RV Jnr which met its end when the tree it was suspended from decide it wanted to lie down...

So I was just playing with bits that were at hand.

This afternoon I had a further play.

First step was to record the measurements from yesterdays configuration of an unknown toroid and speaker wire:

F = 14.2 MHz
R = 96Ω
X = 115Ω
VSWR = 8.5:1

So I then substituted the unknown toroid for a FT140-61 (which arrived in the post).
F = 14.2 MHz
R = 74Ω
X = 103Ω
VSWR = 4.97:1

Finally I substituted RG58U onto the FT140-61

F = 14.2 MHz
R = 56Ω
X = 0Ω
VSWR =1.1:1

While the toroid made a small difference it appears the speaker wire was the real culprit.

Why did I use speaker wire?

It was to hand and numerous "how to's" I have read use speaker and hook up wire.

What is next?

Three things:
1. Have another go with some enamelled copper wire. (I'll have to source some first, I wonder if Jaycar stock it).
2. Have another go with different hook up wire.
3. Try a rod ferrite.

It is amazing how quickly a simple fiddle around in the shack can get out of control.

Again thanks for the advice and suggestions.

[VK3ALB]

Hi Colin,

You might find this interesting.

[VK2AVR]

Certainly highlights the importance of following published designs to the letter, if one expects to get the specified performance out of them.

The "why" is a whole other question - and figuring it out leads to increases in knowledge.

[VK4WDM]

Certainly highlights the importance of following published designs to the letter

I absolutely agree with that statement. I have dived into the junk box to make baluns on several occasions rather that source the published components and ended up wasting time and becoming frustrated. One can assume that the author has done all the fiddling already to get the antenna to work. Then you might have a go at improving things against the original.

Also a plea to authors: make absolutely sure that what you publish is correct and if possible include source and part number then there is no excuse for those following to guess, or use something they think is "close enough."

73

Wayne VK4WDM

[VK3PY]

Colin,

As I understand it, you wound a bifilar winding of speaker wire onto a ferrite toroid. In other words, you produced a common-mode choke using parallel-wire line of unknown characteristic impedance, which, as your measurement shows, wasn't 50 ohms.

Replacing the speaker wire with RG-58 gave you a good match. No wonder - you now have a 50-ohm coaxial line feeding a 50-ohm load. Your measurement is valid in the sense that it tested the match between the transmission line and load, but that's all it did. Take the ferrite core away and you'd get the same result.

What you haven't verified is that your creation provides the desired transformation between a balanced and unbalanced system. So - dare I ask - did you really make a balun?

Chas
VK3PY

[VK4WDM]

When I was in the USA a few years back I brought some high-spec, commercially produced 1:1, 2:1, and 4:1 baluns, and a 9:1 unun made by DX Engineering.

They weren't cheap, but this has proved to be on of my best ever AR purchases. When I am playing with new antennas I use one of them to get the antenna on the air, and then try some HB ones. I can't afford to have commercial baluns on every antenna but the commercial ones are the reference point for experimentation.

73

Wayne VK4WDM