RF Choke, Iron powder core vs Ferrite

Discussion in 'Homebrew and Kit Projects' started by SA7CNG, Aug 4, 2021.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: L-Geochron
ad: abrind-2
ad: L-MFJ
ad: HRDLLC-2
ad: Left-3
ad: Left-2
  1. SM0AOM

    SM0AOM Ham Member QRZ Page

    The loss in the balun is much too high to be explained by core loss alone.

    But an impedance transformation in the coiled-up transmission line may upset the matching sufficiently to create an apparent loss when its length approaches 1/4 wavelength.

    Also, be sure that a 10x probe is used, otherwise the capacitive loading effects will make the results quite unreliable.

    As the power levels are QRP, RG-174 or similar is sufficient at low SWR.

    To find out the common-mode impedance, you can measure the voltage drop over the outer conductor of the coiled coax when connected between the transmitter output and ground with a small resistor (a few ohms) in series.

    First measure the voltage drop over the resistor, and then measure the total voltage over the series connection of coil and resistor. You need to use a 10x probe, otherwise the loading of the oscilloscope input capacitance will throw off the results.

    By first solving for the current by dividing the voltage drop with the low resistance value, and then solving for the common mode impedance by dividing the total voltage by the calculated current, the magnitude of the impedance can be found, with a slight error due to the series resistor.

    In order to be reasonably effective against common mode currents, the impedance magnitude should be at least 4 or 5 times the feedpoint impedance of the antenna.

    N5HXR, KU3X, WA7ARK and 1 other person like this.
  2. WA7ARK

    WA7ARK Ham Member QRZ Page

  3. WA7ARK

    WA7ARK Ham Member QRZ Page

    You have to be careful making this measurement. If the load was an "antenna" (instead of a dummy load), then if there is any CM current on the coax shield before the choke is added, the voltage measured across the "load" with the o'scope will change after the choke is added. This is due to the fact that the antenna feedpoint impedance is different with and without CM current on the coax. The voltage seen by the 'scope changes due to the feedpoint impedance change, and that has nothing to do with loss in the choke!

    In doing the test with a dummy load, you are assuming that there is zero CM current in the cable between the transmitter and the dummy load even before you add the choke. This is a questionable assumption.

    The second problem I see is if there is some residual Swr in the cable between the transmitter and the load (nothing to do with the choke). By inserting the choke, you are effectively adding to the length of the transmission line between the transmitter and the load. Changing length of the transmission line effects the voltage at the load. Even worse, if the characteristic impedance of the wires wound through the choke is different than the rest of the coax used to connect the transmitter to the dummy load, then you are effectively making a Q-section like impedance transformer.

    I have measured proper type 43 ferrite chokes as described by K9YC and G3TXQ, and their insertion loss is so low that it is below my abilty to detect it! Most folks who think that adding a choke to the outside of coax adds loss to the system do not make the proper measurements or misinterpret the results.
  4. SA7CNG

    SA7CNG Ham Member QRZ Page

    I used the Oscilloscopes probes when i meassured the RF-Power, it is a 200MHz oscilloscope, could this be a problem? They were set at 10x
    SWL37632 likes this.
  5. KU3X

    KU3X Ham Member QRZ Page

    You asked, "The impedance in the points you circled is probably higher than 50 ohm, but it is only a few cm, does it still affect the system despite being so short?"

    The answer is YES ! Why do you think the SWR on my balun is so much lower than your's is? I not only used 50 ohm connectors, I kept the shield and center conductor very close together before they connected to the solder points on the connector. I also used 50 ohm coax. Mine will even show a very low SWR up in the VHF range. The -20 DB cutoff for that balun is 75 mhz. From that point on the attenuation drops well under -20DB. So it's not a balun once it gets to 2 meters. But I swept it just for a point of interest.
    I'm not sure what wire you used to make your balun, but if it's #16 enamel, if you look closely at each wrap, I can see a small gap between each wire. It's not much, but it all adds up. One brick here and another one there, before you know it you have an over size load. Hi Hi

    Set your antenna analyzer for 30 mhz and instead of measuring SWR, measure in Return Loss. Put a good dummy load on the output of the balun. Take a reading and then the wires in each circle, put them butted up against one another. Now look at your RL......it drops a little.
    You'll never see a difference when you do this experiment on 80 meters, the wavelength of the wire in question is much too short.
    Barry, KU3X

    Foot note: SWR is to a tape measure what Return Loss is to a micrometer. It's a much more accurate way of measuring.
    Last edited: Aug 8, 2021
    SWL37632 likes this.
  6. SWL37632

    SWL37632 QRZ Member

    Karl-Arne provided very good responses to the question and I agree with him.

    My only add-in to Karl-Arne's inputs is that in my measurements, using a high frequency o'scope probe (150 MHz), the length, placement, and interface means of the ground wire from the probe will give varying results......sufficient as to provide questionable and erroneous readings.

    This "Pigtail" type probe connection needs to be as short as possible, the connection to the probe needs to be solid....many probes use a 'convenience' rotary type of surface interface contact. As large a wire gauge as practical.

    Well....I think you understand where I'm going with this.

    20+ WPM 1970's Extra
    SM0GLD likes this.
  7. SWL37632

    SWL37632 QRZ Member

    My experiments confirm all that Barry has written.

    Remediating each minor appearing thing, will increase the performance.....like Barry's bricks.

    Like your comment:..........."SWR is to a tape measure what Return Loss is to a micrometer. It's a much more accurate way of measuring."


    20+ WPM 1970's Extra
    KU3X likes this.
  8. WA7ARK

    WA7ARK Ham Member QRZ Page

    ...and you would be making a huge mistake!

    I took the model of post #10 and removed the choke. I vary the length of the simulated coax ("variable K", the second "radial") running from the base of the monopole to the transmitter, and varied its length from 10 to 75ft. In post #10, the choke was placed at 17ft from the monopole; now there is no choke there, and the CM current can flow past the 17ft point all the way to the end of the coax (into the transmitter chassis).

    Here is what changing the coax length (no choke) does to the antenna:
    When the choke is there at 17ft down the cable, it effectively makes K=17 in the plots above. There are only two regions where the swr is suitable for use without a tuner in-line.

    This shows the currents when the coax is 35ft long. Note the asymmetry.

    Which shows in the elevation pattern which is skewed in the direction of the resonant radial:

    Here is a detail of what changing the unchoked coax does to swr, Z, and gains:

    Without the choke, the only way this antenna makes sense is if the coax is exactly 17ft long...
  9. G0TNU

    G0TNU Ham Member QRZ Page

    I use QRP transceivers with a 49:1 efhw @ 5w & suffer no problems, Also use doublet antenna with 1:1 balun & LDG Z11 pro with no problems using a very short patch lead.

    Very low power does not cause me any problems.

    Sometimes i attach a counterpoise to the baluns SO239 using a car battery clamp or use a tuned counterpoise.

    I wouldn't use high power for this set-up.

    QRP RULES / CW IS FUN 20210627_095435.jpg
  10. WA7ARK

    WA7ARK Ham Member QRZ Page

    Good for you, but your experience brings no useful information to the present discussion!

Share This Page

ad: wmr-1