Balun measurements

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by VK2FXXX, Feb 13, 2011.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: abrind-2
ad: Subscribe
ad: Left-2
ad: L-MFJ
  1. VK1OD

    VK1OD Ham Member QRZ Page

    Kerry, you are drawing some distinctions that probably don't exist.

    Guanella described a balun made of a transmission line coiled to increase its common mode impedance. It happens that he showed a two wire line, but the maths of it works just the same for a coax as it is the coiling of the common mode 'conductor' that increases its common mode impedance.

    Using a core with permeability greater than one means that less turns are required, and Walt's (W2DU) implementation is just a number of cores with just one turn, or alternatively they could be seen as a cascade of smaller chokes.

    If you examine some of the commercial baluns, they incorporate both techniques, a cascade of several multi turn ferrite cored chokes. This technique is to reduce distributed capacitance.

    My own view is that they are all Guanella baluns, just different implementations.

    BTW, lots has been written about the optimal Zo for line sections in these transformers, including some blatant errors, but it turns out that Zo a little different to the calculated optimal Zo for an idealised transformer may be better as it may partly offset some other effects and give slightly better insertion VSWR at the high end.

    As you note, the nominal 1:1 transformer is only 1:1 (or close to it) for a certain load impedance and over a restricted frequency range. A broadband transformer that delivers on the 1:1 promise independently of load impedance is a myth.

  2. VK2FXXX

    VK2FXXX Ham Member QRZ Page

    Gday VK2TIL
    Whats your name?
    We must have posted at nearly the same time.
    As you say 50 ohm coax is the go for low swr applications or low power .
    Obviously with 10 w as allowed here ,a core with rg174 will be fine even in an atu,Im not sure of the voltage rating if the 174 ,but obviously care would need to be taken with the voltage rise as the load deviates further from 50=j0.
    therefore a balun wound with plastic coated wire in or near the atu,(guanella type) would seem to be the best option for when iI finally upgrade to the 400w

    Im very happy with the bridge ,it is a well put together kit. It does most things I want from it at the moment.
    As you say its not a $10000 vna ,Its a $75 Kit ,and good to learn with.
  3. VK2FXXX

    VK2FXXX Ham Member QRZ Page

    Another question

    Since I dont have a vna.could I set up an experiment where I directly measure common mode current on a length of coax stretched out within easy reach?
    In other words ,purposely set up the coax to have common mode currents ,measure them ,then apply balun and measure again?
    Maybe using a meter as described on w8ji Toms website?
  4. VK2TIL

    VK2TIL Ham Member QRZ Page

    " ... 50 ohm coax is the go for low swr applications or low power ... ".

    No; No; No!

    A 50-ohm line, whether co-ax, twin-line (or clothes-line :) ) is required for a 50:50 (1:1) transformation; it doesn't matter if it's a milliwatt or a megawatt.

    You cannot make a 50-ohm line with twin wires of the order of a mm dia.; Google the equation (or find it in the ARRL book) for Zo of twin-line and plug some numbers in.

    It won't work.

    That's why "old-fashioned" twin line is made in Zo such as 300, 450 or 600; never 50-ohms.

    Small teflon-dielectric cables such as RG-178 or RG-213 can handle a lot of power/voltage; if in doubt go to a larger cable.

    Larger cable is one reason why larger-diameter toroids are used; the larger "ring" is necessary to accommodate the larger cable's diameter and, also, its reluctance to be bent into a small radius.

    Apropos my W2DU/Guanella point; consider that the two conductors inside the co-ax are entirely isolated from the outside world (save for imperfections in the co-axial structure).

    We can bend the co-ax; we can coil it; we can run it along a metal roof; we can put ferrites around it; we will not affect those two conductors.

    So; putting ferrite around the co-ax will not affect it as a transmission line.

    Guanella used twin-line; he was able to coil or "ferrite-load" this line to create common-mode impedance in respect of the two conductors; this is not possible in co-axial line.

    The desirable outcome of Guanella's approach is to pass "up-&-return" (differential-mode) currents while opposing "up-&-up" or "return-&-return" (common-mode) currents.

    With co-ax we can only affect the "third conductor"; the outside of the shield.

    Ferrite-loading of the shield performs the same "blocking" function as a W2DU choke or a ferrite bead on a transistor lead.
  5. G3TXQ

    G3TXQ Ham Member QRZ Page

    You can't quantify the choking impedance of the balun from that sort of experiment unless you can also quantify the common-mode environment in which it is placed; and I would think that is in the range "difficult" to "impossible".

    Steve G3TXQ
  6. VK1OD

    VK1OD Ham Member QRZ Page

    The Guanella 1:1 balun works by creating an impedance to common mode current that does not affect the differential current.

    In the case of a pair of wires, the common mode current (which is the sum of the currents in each wire, having regard to magnitude and phase) gives rise to magnetic flux which is concentrated by techniques such as coiling the conductor, using a high permeability core, so increasing the common mode impedance. (The differential components of current cancel each other's effect a very short distance from the wires.)

    In the case of TEM mode coaxial cable where skin effect is fully developed, a property is that the current on the outside surface of the inner conductor causes an equal but opposite current on the inside surface of the outer conductor. This is purely a differential current (by definition), and so any common mode current at the two terminal end of the line MUST flow entirely and exclusively on the outside surface of the outer conductor. External fields are due only to the common mode component of current. Again, in a Guanella balun using coax, the common mode current gives rise to magnetic flux that is concentrated by techniques such as coiling the conductor, using a high permeability core, so increasing the common mode impedance.

    You could argue that the differential components of electric and magnetic fields of a two wire line are partially immersed in the core material, but the effects are so small as to be insignificant. At the best, the core material reduces magnetic reluctance by a tiny amount as it is far from a closed magnetic path, and similarly, the permittivity of the core increases electric flux by only a tiny amount because only a tiny amount of electric flux is immersed in the core. This can be verified by measuring Zo for the line before and after winding it onto a core, there is usually insignificant change.

    A balun can be made with two wire line or coax, with or without a magnetic core, one core or many, one turn or many. What makes it work is the common mode impedance that is created, and the effect of that on common mode current as the lower is common mode current, the more perfect is current balance. When the physical length of the choke is small wrt wavelength, the lumped component approximation of a lossy inductor with some equivalent shunt capacitance is a good approximation.

    Until now, I have not seen anyone argue that since Guanella described a balun that used a pair of wires and did not use a magnetic core, that baluns that use coax or a magnetic core are not Guanella baluns.

  7. VK2TIL

    VK2TIL Ham Member QRZ Page

    This bloke is thinking!

    Certainly better than I am at the moment; I'm a Senior Citizen and it's well-past my bed-time. :)

    But consider a co-ax line with a generator at one end and a dummy load at the other.

    Now put a resistor, perhaps 50 ohms, from the co-ax centre to the outside of the co-ax; voltage-divider effect will direct a quantifiable current from the centre conductor to the outside of the shield.

    Measure that current with & without the balun.

    I will have to think about this overnight; Matron is calling at present. :)
  8. VK2FXXX

    VK2FXXX Ham Member QRZ Page

    Kerry, my point was that using rg178 coax wound around a toroid in an ATU would not be wise ,if using 400w cw into an extreme load.Is this correct?
    However such a balun at the feed point ,of a coax fed 60ohm feedpoint dipole at 100w ,would be good.
    This raises the argument ,of which side of the ATU to put the balun.

    Yes ,understood,Owen explained the impedance transformations I measured very well in an earlier post.

    Yes but winding RG213 on a 35mm diameter toroid may be difficult ;) As you say later in your post, a larger toroid is needed for the thicker cable.This is why the bi-fillear (spelling) winding method makes sense in an ATU. No need for making the transmission line section through the balun 50 ohm.
    The coax through the toroid method as you say is best for 50ohm-50ohm Baluns.

    Kerry please do have a think about the common mode experiment and get back to us.

    Steve ,
    Ok ,I remember reading an article ,I think it was one of Toms, where he made a meter ,calibrated to 1amp I think.This meter was made of very little metal parts, Its use was to measure the current on the outside of coax fed antennas.
    It was designed to disturb the currents as little as possible.
    the article :

    My Idea was to create common mode currents on a coax feedline at least a 1/4 wavelength long,maybe 1/2 wl .This feedline would be within easy reach of the tester,with the meter on a non conductive stick.Measuring current and /or voltage with and without a balun.
    Are you saying that such a test would not be valid,as the feedline and antenna are not in a `normal` environment.

    I would think that having control of the environment,would make quantifying it easier?
    How do I measure a balun doing its job?

    Owen your post has helped my understanding of baluns a great deal ,and as usual for your posts and articles ,I will have to read it several more times at a civil hour to digest it all.
    Im off to bed.
    Thanks for the discussion.
    Last edited: Feb 13, 2011
  9. G3TXQ

    G3TXQ Ham Member QRZ Page

    The OP's proposed experiment would be something akin to this:

    I have a torch. I notice that if I place an unmarked resistor in series with the bulb the brightness reduces by 90%.

    What is the value of the unmarked resistor?

    Steve G3TXQ
  10. VK1OD

    VK1OD Ham Member QRZ Page

    I don't really understand your test, but the common mode current in a current balun deployed in a multiband HF dipole for example is not simply related to the differential load impedance at the point of deployment.

    I know lots of people use ohms law to calculate the common mode current, but I put to you that they are wrong. The common mode current is typically a standing wave on the common mode conductor and can be calculated by MOM modelling of the entire antenna system (ie including the common mode conductor).

    That is why I said that bench tests of the balun (in respect of power handling) are not good indicators of how the balun will perform in antenna system service. It might also explain the 5kW rating of some commercial baluns.

    Sure it is easy to do ohms law calcs for an interstage balun in a box of electronics, the approximation will usually be good, and it might be good for a current balun at the feedpoint of a Yagi under some conditions.... but not in the general case for an antenna system.

    The article Baluns in antenna systems discusses the flaws in some of the simple models often used.

Thread Status:
Not open for further replies.

Share This Page