Results 1 to 4 of 4

Thread: A novel way of broadband matching a small loop HF antenna

  1. #1

    Join Date
    Sep 2019
    Location
    San Marzano Oliveto, Asti, ITALY
    Posts
    7

    Default A novel way of broadband matching a small loop HF antenna

    METHOD FOR PASSIVE BROADBAND MATCHING OF A SMALL LOOP ANTENNAS IN HF

    NOTE: this work was started on the SDRplay Community Forum

    https://www.sdrplay.com/community/vi...=4698&start=20

    but I could not complete publication because of Forun closure. I am restarting the publication here.

    PART 1

    A single turn loop antenna is very simple to build. For the development shown below I have made a square loop with a side length of 88 cm, using wood planks from a discarded pallet. The advantages of a square loop are its small dimensions, better immunity from local man made noise and practically no cost. The drawback is that the antenna equivalent circuit is a resistor in series with an inductor and to get signal to the receiver you need (A) either a very high impedance termination, to avoid voltage drop on the equivalent series inductive reactance or (B) a parallel capacitor to resonate the inductance, but in so doing you lose the broad band capability. Solution (A) is broadband, but requires the use of a wideband amplifier up front, which can overload and generate second and third order intermod products in receive and make the antenna useless in transmit.

    The proposed method is to use a special broadband matching circuit which makes the troublesome antenna series inductance disappear, without resonating the system.

    This method is based on original work by Philip R. Geffe (Simplified Modern Filter Design) who, as far as I know, was the first (in the fifties) to publish the idea that filter networks can be used not only for frequency band limiting, but also for broadband impedance matching. In his book Geffe gives tabulations for normalized elements of filter neworks working between different input / output impedances. Far more complete tables have been published later by Saal and by Zverev. For complete reference information, look here:

    1. Handbuch zum Filterentwurf by Rudolf Saal, AEG-TELEFUNKEN, Berlin, 1979.
    2. Simplified modern filter design, by Philip R. Geffe, John F. Rider Publisher, New York, 1963.
    3. Handbook of Filter Synthesis, by Anatol I. Zverev, John Wiley & Sons, Inc., Hoboken, N.J., 1967
    4. Normalized filter design tables, taken from Filter design handbook and downloadable as PDF here:
    http://www-3.unipv.it/lde/didattica_...signTables.pdf

    NOTE 23/10/2019: corrected reference 4 above, now it should work! Apologies for previous reference not working.

    Looking at the equivalent circuit of a small single turn loop, we see that it is represented by a resistor of a very low, sub-Ohm value, in series with an inductor. If we consider an useful frequency band extending from 3 to 15 MHz, we see that the median values are R=0.56 Ohm and L=4.6 uH and that the spread is relatively small. It must be noted here that I calculated the loop inductance with the formula shown below(*)

    Now my idea is very simple:
    a) if I take a simple Butterworth or Chebyshev prototype, say N=5 and
    b) de-normalize it @ 15 MHz (or any frequency above the reception range I want from my loop) and
    c) also de-normalize between 0.56 Ohm input and a convenient higher output impedance and
    d) if the first series inductor has a calculated value higher than the antenna inductance,

    THEN

    e) I can incorporate the loop inductance into the value of the first network inductor and
    f) Obtain perfect wideband match between 3 and 15 MHz without resonating the loop.

    I shall now test the idea by calculation and if there is a significant number of simple networks that fulfill the conditions described above, I shall show them next.

    (*) Inductance L = 0.008 * s * ln(s/l + 0.37942 + (0.333 * l/s)) = 4.29569 uH

    Where:
    L: inductance in uH
    s: side length, cm = 88
    l: coil length = wire diameter cm = 0.2
    Last edited by glovisol; Wed 23rd Oct 2019 at 17:47. Reason: Reference corrected /duplicated text/reference corrected again

  2. #2

    Join Date
    Sep 2019
    Location
    San Marzano Oliveto, Asti, ITALY
    Posts
    7

    Default PASSIVE BROADBAND MATCHING OF A SMALL LOOP ANTENNAS IN HF - continued

    PART 2

    NOTE: The theory of the proposed method becomes very simple in practice, when explained by a few schematics/illustrations, which I am not allowed to post at the moment, so I shall add them in future posts, when necessary permissions will be available.

    Loop antenna equivalent circuit values in the HF frequency have been confirmed by MAMMANA antenna modelling performed by Phil, VK7JJ: note that the conductor diameter was set at 2.5 mm and we have a "real part" of Z which is the sum of radiation and of loss resistances. Calculated values are as follows:

    Wavelength...........................80m.......... ......40m.............20m
    Frequency fo, KHz..................3,550...............7,150.... ......14,150
    Z, real part Ohm....................0.38................0.56... ..........1.1
    Z, im part (inductive) Ohm......99...................207..............451
    Inductance @ fo, uH...............4.4..................4.6......... ......5.1
    Resonating capacitance, pF......453.................108...............25

    The method outlined in the previous post is indeed theoretically possible, using data for polynomial filters with unequal terminations. In Reference 3, previous post, tables of lowpass element values are presented for filters accommodating not only in/out impedance ratios up to 1:10, but also prototypes matching a given impedance to open circuit (infinite Ohm) or to zero Ohm. This property is essential, because broadband loops made with thick conductors have real part impedances measurable in milliOhms, so the matching circuit must accommodate a termination approaching a short circuit.

    In order to have a really low impedance loop, I calculated the parameters for a loop antenna with the same dimensions as the square loop specimen, but with much thicker conductor, bringing the diameter up to 15 mm, using the Magloop calculator spreadsheet Ver. 1.23. With this spreadsheet calculator we get the value of the radiation resistance separate from the loss resistance, so it is possible to optimize loop parameters before actually building the prototype. Increasing the diameter drastically reduces inductance and radiation resistance, while at the same time increasing potential bandwidth. Results are shown below: please note that resistances are now given in milliohms.

    Wavelength.................................80m.... ............40m.............20m
    Frequency fo, KHz........................3,550...............7,1 50..........14,150
    Radiation resistance, mOhm..........0.5..................7.............. ...110
    Loss resistance, mOhm.............:...34...................47...... .........66
    Inductance @ fo, uH.....................2.8..................2.8... ............2.8

    In the next post we shall look at a possible broadband matching circuit.
    Last edited by glovisol; Wed 23rd Oct 2019 at 09:34.

  3. #3

    Join Date
    Sep 2019
    Location
    San Marzano Oliveto, Asti, ITALY
    Posts
    7

    Default PASSIVE BROADBAND MATCHING OF A SMALL LOOP ANTENNAS IN HF - continued

    PART 3

    The simple loop antenna made with wood strips salvaged from a pallet and previously mentioned is shown below:

    Cross loop antenna .jpg


    Its series inductance of 2.8 uH, can be embedded into a low pass matching network such as the one shown below:

    Low pass impedance xfrmr.jpg

    As seen in the schematic, the computed low pass filter network has L1 with a value of 2.8 uH. The matching network requires L1= 3.46 uH, so by using L1=0,66uH we effectively make the loop inductance disappear without resonating it! In practice one terminates the loop with a very small coil.

    This matching network matches 200 Ohm to the short circuit (extremely low impedance) represented by the antenna. Then the wideband transformer T1 lowers the impedace level to the standard 50 Ohm value, or brings it up to any desired value. If using the SDRduo receiver's balanced Hi Z input, then T1 should have a 1 KOhm secondary.

    The attached image shows the calculated equivalent circuit of a small loop antenna made with thinner wire and hence with higher equivalent series inductance, but the concept remains the same.

    In Part 4 next we shall see matching network details and frequency response.
    Attached Images Attached Images
    Last edited by glovisol; Wed 23rd Oct 2019 at 17:44. Reason: Text incomplete

  4. #4

    Join Date
    Sep 2019
    Location
    San Marzano Oliveto, Asti, ITALY
    Posts
    7

    Default PASSIVE BROADBAND MATCHING OF A SMALL LOOP ANTENNAS IN HF - continued

    PART 4

    Here below is the spreadsheet calculation and frequency response of the broadband matching antenna network

    0-200 Ohm match graph.png

    0-200 Ohm match.png

    A more robust antenna can be made by adding the sides of the square to the cross members, as shown below.

    129 cm loop.JPG

    I shall now test circuit and antenna and report results next.
    Last edited by glovisol; Wed 23rd Oct 2019 at 17:46.

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •