Coax Loop Balun for 6M Moxon

[VK6ZFG]

I was looking for a balun for a new 6M Moxon antenna I have in mind to build and decided to give a coax loop balun a try to use with it.

I made up a coax loop balun and after adjustment using a tracking generator it finished up being 6 turns of RG58 zip clipped together in about a 45mm diameter.

The outer of athe coax combined with stray capaclty forms a parallel tuned circuit. This is something that seems to be overlooker in coax balun loop articles I have read. For best results from the coax loop balun resonance needs to be tuned for the frequency of operation required.

Coax Loop Balun Sweep

A possible option might be to use slightly less coax in the loop and have a a small capacitor trimmer between the ends of the coax loop to fine tune the balun.

[VK6OX]

Interesting Igor.

When I first became re-acquainted with 6m here in Perth, I built a lightweight 5-el DK7ZB yagi and IIRC the "choke balun" I wound was very similar to the dimensions (# of windings and dia.) as yours.

I might try and replicate your balun, but use my SARK-110 as the swept signal source/display.

Stay tuned!

[VK6ZFG]

Andy

You might find that if the loop is wound flat on a coil former, rather than bunched as was in my case, it may exhibit better isolation at frequencies higher than the resonance point due to the stray capacity being lower hence bypassing the inductance less at higher frequencies.

From what I can see the coax loop balun is not really a broadband device as is often made out to be, but rather a specific frequency device with somewhat reduced performance at other frequencies.

[VK7HH]

I was looking for a balun for a new 6M Moxon antenna I have in mind to build and decided to give a coax loop balun a try to use with it.

I made up a coax loop balun and after adjustment using a tracking generator it finished up being 6 turns of RG58 zip clipped together in about a 45mm diameter.

The outer of athe coax combined with stray capaclty forms a parallel tuned circuit. This is something that seems to be overlooker in coax balun loop articles I have read. For best results from the coax loop balun resonance needs to be tuned for the frequency of operation required.

50mhz loop balun 1.jpg

A possible option might be to use slightly less coax in the loop and have a a small capacitor trimmer between the ends of the coax loop to fine tune the balun.

What method did you use to measure your coax choke?

Tracking generator with the coax choke in series?

[GM3SEK]

I was looking for a balun for a new 6M Moxon antenna I have in mind to build and decided to give a coax loop balun a try to use with it.

I made up a coax loop balun and after adjustment using a tracking generator it finished up being 6 turns of RG58 zip clipped together in about a 45mm diameter.

The outer of athe coax combined with stray capaclty forms a parallel tuned circuit. This is something that seems to be overlooker in coax balun loop articles I have read. For best results from the coax loop balun resonance needs to be tuned for the frequency of operation required.

A possible option might be to use slightly less coax in the loop and have a a small capacitor trimmer between the ends of the coax loop to fine tune the balun.

Most people call this a "choke balun" or "current balun". Almost everyone will have opened this thread expecting to read about a 4:1 step-up balun using a half-wave loop of coax, because that's what the term "loop balun" usually means.

The resonant frequency that you measure will depend on the distributed capacitance of that specific method of winding the coil - but it will also depend critically on the stray capacitance of your test setup. This parallel resonance is high-L but low-C, so even one picofarad of uncompensated strays can move the measured resonant frequency by several MHz. I don't know any accurate way of compensating for this, short of using a Vector Network Analyser with the full Open-Short-Load calibration method and also a lot of care about not moving the test leads between the calibration and the measurement.

Also the self resonance of an air-wound coil is quite narrow, and very easily detuned when the choke is installed on the antenna. So even if you did measure the resonant frequency correctly on the bench, the resonant frequency up in the air is guaranteed to be in the wrong place!

These points have been extensively discussed in articles by G3TXQ, K9YC and myself... which is why all three of us recommend using a ferrite-loaded choke baluns. The reason for adding ferrite is to create a high resistive component to the impedance of the choke, because that resistive component cannot be detuned.

This has all been well worked out for HF, where the best solutions use multiple turns on a toroid to make the parallel resonance cover the bands of interest. VHF is more difficult because of the measurement problems but it's on the list of things to do next. For the moment, a string of ferrite beads looks like a fairly effective solution for VHF (multiple turns on a toroid would take the resonant frequency too low).

The final question of course is "Who needs all this? A coil of coax is plenty good enough for me!"

Well, that's probably because the feedpoint of a Yagi is already quite a well-balanced load, so the balun doesn't need to be spectacularly good - just good enough to avoid messing things up . That's probably the main reason why several different types can all seem "good enough".


73 from Ian GM3SEK

[VK7HH]

Almost everyone will have opened this thread expecting to read about a 4:1 step-up balun

I didn't. I searched for "choke balun"

Well, that's probably because the feedpoint of a Yagi is already quite a well-balanced load, so the balun doesn't need to be spectacularly good - just good enough to avoid messing things up . That's probably the main reason why several different types can all seem "good enough".

There seems to be a lot of talk of coax lengths vs number of turns etc. I wonder where they calculate all this out let alone measure it.

If I wanted to build one of these antennas -
http://vk2zoi.com/articles/half-wave-flower-pot/

How do I scale up/down the coils at the bottom of the antenna to suit different frequencies, for instance 52MHz on a 32mm former and know that it is working, especially with such a narrow self resonance. Similar too with a choke for a Moxon, dipole, yagi etc. Best return loss?


Ferrites available from Jaycar -

http://www.jaycar.com.au/productView.asp?ID=LF1292

These state typical impedance at 25MHz 300ohms approx, 590ohms at 100MHz dropping off after that. But then again others have said they have had no success with clamp ferrites, plus they play up if you get water in there.

[VK6ZFG]

In Australia cars have mudgards and boots but when the same cars are sold in the USA have fenders and trunks.

The coax 4:1 balun I believe should be described as a "4:1 half wave coax loop balun" as the presence of the half wave of coax is an essential part of the device. This half wave makes it work and makes it frequency dependant. The length of is not arbitary but critical to its successful operation.

My preference for baluns is to make them using ferrite toroid or beads and it is what I normally use.

Coax loop baluns have been around for a long time but I have always had questions as to how well they perform. The builting of a new antenna motivated me to check one out. The test was carried out using a Rigol DSA815TG in tracking generator mode with the device normalised before the coax loop was inserted. Fortunately the input and output ports were about the right distance apart so this worked out well. Attenuator pads were not used in series input/output ports as ultimate measurement accuracy was not being persued

A coax loop balun essence creates an inductive choke in series on the outer braid of the coax (but not inside of thee braid). The test confirmed that the isolation provided was not all that great. However when the resonance frequency is reached the isolation provided greatly improves as the test indicated. Making use of this resonant point in practice a challenge. Fortunately with a 50 ohm feed point antenna the task is difficult but achievable.

It is a different story if the antenna impedence is high as is the case at the end of a dipole. This requires the isolating device to provide a very high impedance. This is where I have problems with designs such as the flower pot antenna. The number of coax turns used on the flower pot antenna to me indicates that the resonant point is likely to be well below 2 metres. It is as a result it is unlikely to provide resonance isolation and even if it did provide the much higher isolation impeance required. I also would also expect that the top end of the dipole to be shorter than the bottom half due to the velocity factor of the covered wire.

All antennas will radiate to some extend and can be adjusted to exhibit a low VSWR but VSWR is not a measure of performance as a dummy load has a very low VSWR but is a very poor radiator.

[VK7HH]

My preference for baluns is to make them using ferrite toroid or beads and it is what I normally use.

How do you make them? What beads do you use?

It's difficult to get multiple turns of say RG-213 through some of the clamp on types. How much impedance do you require for VHF and above?

[VK6ZFG]

I have a pile of ferrites bought and many salvaged from all sorts of devices. Problem is you need to evaluate them until you find one that suits.

RG213 is a problem. Why not use a a thinner coax just for the balun and insert it in series are the desired point in the feeder? What I usually do is include the balun in a box. In the case of VHF/UHF yagsi etc the box can also serve as the mounting point for the driven element. Insulated cable glands can be used to permit the driven element to exit the box and Stauff clamps internally can provide additional support.

[img]

[/img]

[VK7HH]

Problem is you need to evaluate them until you find one that suits.

Hi Igor, Yes. Seems to be a common trend with these things. How do you know when they suit? Lowest SWR?

RG213 is a problem. Why not use a a thinner coax just for the balun and insert it in series are the desired point in the feeder?

That is possible. I guess you need to watch the losses at UHF. You don't want to be using anything like RG58. LMR-240 type has a solid centre conductor, which makes it difficult to coil.

What I usually do is include the balun in a box.

I don't really do boxes. Condensation can build up inside and especially so if there is no drip or weep holes in the box.
RTV Silicon sealant on the feedpoint and over the coax seems to be the best method. Your photo is a little hard to see where the toroids are. I assume they are under the heatshrink right before the feedpoint or is it a clamp one I cannot see with the multiple turns through it.

[VK3MIX]

Insulated cable glands can be used to permit the driven element to exit the box and Stauff clamps internally can provide additional support.

Where are you sourcing your stauff clamps from?

[VK6ZFG]

The earlier photo was for the coax loop balun I am currently playing with what will be a secondary antenna.

Photo below is for my 6M 5 ele yagi that has been in service some time but not yet back in service. The balun uses ferrite. Actually has two baluns inside the box (metal in this case).

6M DE Loop

If using sealant, do not use Silastic as this is acidic and will cause corrosion (as one reputable commercial industry antenna manufacturer found to their expense). Use gutter sealant instead.

I purchase the Stauff clamps locally. They are used for pipe supports and are available from outlets that supply industry. Do a search on the internet.

One has to be realistic about cable loss. The length (relates to frequency) and loss of the coax required to make a balun is insignificant compared with the feeder length and loss.

[VK4TI]

Whttp://www.stauff.com.au/index.php?id=1423 , bread board works well too

[VK6ZFG]

have a look at:

www.stauff.com.au

they list suppliers

[GM3SEK]

In Australia cars have mudgards and boots but when the same cars are sold in the USA have fenders and trunks.

The coax 4:1 balun I believe should be described as a "4:1 half wave coax loop balun" as the presence of the half wave of coax is an essential part of the device. This half wave makes it work and makes it frequency dependant. The length of is not arbitary but critical to its successful operation.

If "loop" is a good Australian word for a multi-turn coil, then I apologise

Back now to the technical stuff:

My preference for baluns is to make them using ferrite toroid or beads and it is what I normally use.
Coax loop baluns have been around for a long time but I have always had questions as to how well they perform. The builting of a new antenna motivated me to check one out. The test was carried out using a Rigol DSA815TG in tracking generator mode with the device normalised before the coax loop was inserted. Fortunately the input and output ports were about the right distance apart so this worked out well. Attenuator pads were not used in series input/output ports as ultimate measurement accuracy was not being persued

A coax loop balun essence creates an inductive choke in series on the outer braid of the coax (but not inside of thee braid). The test confirmed that the isolation provided was not all that great. However when the resonance frequency is reached the isolation provided greatly improves as the test indicated. Making use of this resonant point in practice a challenge.

It certainly is a challenge with an air-wound choke because the parallel resonance is so narrow, and so easily detuned by changes in stray capacitance. Even if the resonance is on-frequency in the measurement setup, it will almost certainly go off frequency when installed on the Yagi. Even worse, the choke will be detuned by a further unknown amount when the feedline is connected.

Fortunately with a 50 ohm feed point antenna the task is difficult but achievable.

Agreed. Also a Yagi has a well balanced feedpoint in the first place, so when we connect the feeder the balun doesn't have to 'improve' anything at all - it's mostly about not messing things up! That shouldn't be too much to ask of any feedline choke, and in most cases an air-wound choke will be good enough to prevent common-mode current flowing onto the outside of the feedline.

We can still mess up the balance ourselves, of course - for example by taping the choke onto the metal boom, or by installing the choke inside a metal box. Both of those will create a one-sided capacitance between the driven element and the boom (because the coax shield is connected to only one side of the feeedpoint and not the other). So adding a one-sided capacitance will upset the balance and allow more common-mode current to flow... oops. But that wouldn't be the fault of the choke, only the way it has been installed.

It is a different story if the antenna impedence is high as is the case at the end of a dipole. This requires the isolating device to provide a very high impedance.

Again, agreed. All the problems become much worse when the impedance at the choke point is so much higher. An air-wound choke can have a high impedance, but the chances of getting that narrow resonance exactly on-frequency are effectively nil. The unknown length of feedline is going to detune the choke, so the outside of the feedline will then become an active part of the antenna. Even if it works well enough for some purposes, it's still ugly.

That is why we both agree that ferrite chokes are better. The ferrite loading creates a broader and more resistive impedance, so even if the resonance shifts when you install the choke (and it will) you'll still have a high enough impedance to suppress the common mode current.


73 from Ian GM3SEK

[VK6ZFG]

Hi Ian GM3SEK

Looks like we are both sailing in the same direction under different winds.

The difficulty in achieving resonance on the desired frequency was evident during my measurements. This is consistent with your findings. The resonant null was initially set up to be on exactly 50.000MHz. However by the time the screen was saved it had moved slightly higher (as can be see on the screen shot). The ability to reproduce a balun with the same resonance point by simply by following written instructions is an unrealistic expectation especially at VHF/UHF.

Scaling to another frequency band is even more difficult. It can get you in the right direction but that is all. Does anyone know how to scale a tuned circuit to an exact alternative frequency without the need to fine tune it? I doubt it.

The concept of using the resonance point to provide isolation is easy to grasp. The practical implementation is somewhat more difficult especially if one does not have access to suitable test equipment.

The implementation must also be appropriate for the intended task. If the intent is to isolate a high impedances then the system used must be able to provide an even higher impedance. Tuned circuits have finite impedances depending on how they are made. They do not have infinite impedances.

[VK6ZFG]

Just did some tests using six LF1258 ferrite cores (from Jaycar) on a short length of RG58 coax. Results below:

6 x LF1258 Ferrite Cores in Series on RG58

The ferrite core balun as tested does not have a peak isolation as good as that for the coax loop, but only if the installed loop actually finishes up the desired frequency.

The ferrite core approach has no critical tuning requirement as does the coax loop balun. This means that the ferrite core balun is readily reproduced. It avoids the need for any measurements (and associated equipment).

The ferrite core balun is the way to go. If higher isolation is required one can always add additional ferrites.

[VK3UDX]

Great work Igor testing this, was going to run a few tests myself, you've done the hard work for us all, clamp on ferrite cores compared to coax loop balun has great physical advantages keeps feed line neat in a stacked array and avoids a nice perch for birds hi hi

Would like to run some tests on higher bands time permitting

73 Geoff
VK3UDX

[GM3SEK]

Just did some tests using six LF1258 ferrite cores (from Jaycar) on a short length of RG58 coax. Results below:

The ferrite core balun as tested does not have a peak isolation as good as that for the coax loop, but only if the installed loop actually finishes up the desired frequency.

The ferrite core approach has no critical tuning requirement as does the coax loop balun. This means that the ferrite core balun is readily reproduced. It avoids the need for any measurements (and associated equipment). The ferrite core balun is the way to go.

Thanks for the new measurements, Igor.

The smaller value of peak isolation (or peak impedance) is probably OK for this application, because the feedpoint of a VHF Yagi is already quite well balanced. When we connect the coax, our only aim is to avoid upsetting a good situation that already exists, and we can probably manage that with quite a moderate value of common-mode impedance.

If higher isolation is required one can always add additional ferrites.

Probably... or, well, maybe. The existing broad resonance is already a long way below 144MHz and extra beads might move it even lower.

Another way to enhance the common-mode impedance without adding more beads is to leave a quarter-wavelength loop of coax hanging in free air before the the coax comes back to run along the boom. Some people then like to "ground" the coax shield to the boom (eg by fitting a connector on a metal bracket) but I'm not sure if that is a good thing, because there is no true "ground" up there.

The fashion in Europe seems to be moving towards one or two beads at the feedpoint followed by the quarter-wave loop... but the true common-mode impedance is almost impossible to measure at 144MHz.


73 from Ian GM3SEK

[VK3AUU]

The savvy amateur doesn't use lossy RG58 to make baluns at VHF.

David