Effects of complex impedance thru a current transformer

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by W4KJG, Apr 23, 2016.

ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: Left-3
ad: abrind-2
ad: L-MFJ
ad: Left-2
  1. KA8NCR

    KA8NCR Ham Member QRZ Page

    Agree completely, far too often people fail at basic sentence structure. Behold this gem posted near the top of this thread: "They can do it them self"

    Ouch! This country is screwed.
    W9AMM likes this.
  2. W9AMM

    W9AMM Platinum Subscriber Platinum Subscriber QRZ Page

    I see what you did there. :D
  3. AE7XG

    AE7XG Ham Member QRZ Page

  4. VK2TIL

    VK2TIL Ham Member QRZ Page

    Yes; I made a basic error in relation to the capacitive loads. The expected result is 4C, not C/4.

    The transformer actually did a good job of transforming those loads, at least at frequencies away from the resonance; I apologise to any transformers that may have been offended by my remarks.

    So resonance is an undesired result of capacitive loads; perhaps Ken, the OP, encountered this in his measurements.

    Apropos inductive loads, leakage inductance in a transformer is modelled as an inductance in series with one winding; that is effectively the case when an inductive load is presented to the transformer as in my test and may, therefore, account for the not-perfect test result in this case.

    G'day 'UAQ; the transformer is as depicted in my post;

    I used a small 4:1 transformer that I had lying-around, ten turns twisted on an FT50-43;


    and connected in this manner;


    Test loads were connected to the "4R" side and the impedance at the "R" side was measured.
  5. WB2UAQ

    WB2UAQ Ham Member QRZ Page

    Thanks for the additional info. I signed in at work and I can't see attachments or pictures on QRZ any longer for some reason.
    Would the transformer you are using be considered an auto transformer? Sevick and others have confused me with the way they depict transformers. I think I uploaded a hand drawn schematic diagram showing the 1:4 transformer I was thinking about. Because the coupling is not thru the core may be the impedance will be transformed better or worse than the conventional transformer. I will do the same measurements as you did with this type. Might have to wait until Sunday before I can get to it. I'd like to use the Guanella type as well where there are two bifilar windings on separate cores.
    With the single bifilar winding 1:4 TLT there is a phase delay on one side. With the 2 bifilar windings on separate cores type there will be a short transmission line delay but impacting both legs. Hope that made sense:)
    73, Pete

    Attached Files:

  6. VK2TIL

    VK2TIL Ham Member QRZ Page

    I use Postimage to put photos in posts but your work must not like outside photo-hosts; I will try direct attachment.

    The transformer I used is the simple auto-transformer type; I use them in small amplifiers and usually have a couple lying-around.

    Transformer Diagram.gif Test Setup 2.JPG

    I don't know that there's a lot of transmission-line behaviour going-on in these; I find that their upper limit is around 40MHz or so which is the frequency at which the u' of the #43 ferrite has dropped to about 10% of its ui and has really ceased to be effective.

    I think that the close-twisting approaches the impossible ideal for a conventional transformer where each winding occupies the same space and is therefore cut by the same magnetic lines-of-force.

    Experiments with other transformer topologies would certainly be interesting.
  7. W4KJG

    W4KJG Subscriber QRZ Page

    [I haven't been ignoring this thread. My Windows 10 ham shack computer had a serious brain stroke over the weekend.

    On Monday morning, it came up with a new background I'd never seen before. It came up. and I was able use it for most functions. It unfortunately would not recognize my SDRs via SDR-Radio Pro xxx, SDR-Sharp, or even the old HDSDR. I re-ran ZADIG, and they reluctantly began to work, but I'd lost all my configurations. I also lost my audio links between my SDRs and programs like WSJT-X and FLDGI. These programs were still working through my analog transceiver interfaces.

    Then I lost mouse and keyboard control.

    So I'm presently reinstalling Winodws 10.

    At least I have good backups of most of my stuff.]

    Back to transformers and impedance issues.

    First, I have to really thank several people, especially VK2TIL and KD6RF for their input.

    My issue turned out to be stray capacitance and leakage inductance, in a situation where I expected the SRF (Self-Resonant Frequency) to be about an octave higher than where I was using a 9:1 trifilar transformer.

    I did some of experimenting after this episode by changing to a toroid core 1/3 the original size, moving from teflon coated wire to enameled wire, and hard-mounting the transformer on a small square of G-10 epoxy PC board. At that point, things became much closer to what I expected.

    Based on what I'm trying to accomplish, I've abandoned this original idea and design.

    I mostly operate QRP. Yet, I was concerned about going to the much smaller core and smaller wire. There are times when I might want to go up to 25-50 watts. It caused me to move to a resonant transformer design that seems to be far more stable. Like my looks, these antennas will mostly be mono-band.

    I also found a lot more information -- mostly put out by our Aussie and Brit friends. It really helped a lot.

    KL7AJ -- "This can usually be done with a Smith chart....if you make the assumption that the transmission line is effectively 1/4 wave. The sign of the reactance will be flipped."


    I know I'm getting old and set in my ways. As an oldztymer, I can still enjoy plotting a Smith Chart by hand. I still have my late-60s 8x8 inch "rocker-style" Smith Chart rubber stamp that I use on 8-1/2 x 11 paper to put down a nice round 8-inch Smith Chart graph outline. I mostly use a noise bridge and either a battery-operated Sangean synthesized SW radio, or an RTL SDR dongle/notebook to determine R +/- j to design broad band matching networks for my home antennas. I also do this because I can't afford the types of network analyzers I had access to before I retired.

    I'm old enough to know that things are not always as they appear at first glance, especially when it comes to matching broad band impedance with transformers.

    If it was as simple as finding my septic tank with a pair of copper "witching rods", some of us might be more well-to-do, and my cell phone would work more reliably here in the mountains.

    I'm glad you are all here. It was easier when I was still working in a moderate size engineering department, when these kinds of things were worked over with donuts and coffee in the lunchroom with some really brilliant friends.

    I ain't sure if I gots any grimmatical or speliing aerors in this post, but yoo gits watts yoo gits!

    West Virginia Ken
  8. WA7DU

    WA7DU Ham Member QRZ Page

    "When people can't learn they get frustrated and get sort of crazy. That's what I see here."

    That sentence was lifted intact from your diatribe. It reads as if it is a self diagnosis, and darn accurate in my judgement.

    Let's have a look at your curriculum vitae. I, for one, don't think you are the hot-shot you think you are. If you are employed by any of the advanced technology companies in the U.S., I predict that you will not go far as a valued employee. Your deficits are obvious (a dreadful personality, for one). That is, unless you are a Jeckyll and Hyde, who puts on a different face when meeting with people within punch-in-the-face distance. Seriously, post that CV here. If I am wrong about you, I'll publicly apologize. Of course, the CV should have at least three professional reference letters for validation.
    Last edited: May 1, 2016

Share This Page