[Elecraft] [K2/KX3] Tuning a magloop

[Gary Hawkins]

Dave I read your post with interest, as over the weekend I was looking at an automatic loop tuner described in the November/December isse of ARRL QEX Magazine.  This was for a MFJ-1788 Magnetic Loop working with a KX3 and based on an Arduino micro-controller.  I don't own the MFJ product but I do have an Alex Loop and with a few simple modifications the circuit should work nicely.

The loop tuner uses a DC motor but I thought a stepper might be more elegant.  Thus, I needed to determine the working bandwidth of the Alex Loop to get some idea of the angular resolution I needed from either a direct drive stepper, or a stepper using gearing.  I looked at the Alex Loop around 7, 18 and 30MHz.  3:1 VSWR bandwidths were measured using a RigExpert AA-54 antenna analyzer.  For 3:1 VSWR the measured operating bandwidths at these three frequencies were determined as 44, 65 and 190kHz respectively.  Pretty narrow at the lowest frequency.

Thus, to get a good chance of achieving a VSWR approaching minimum, I figured the angular motion should be fine enough at worst case (7MHz)to achieve an angular motion of the tuning capacitor of no more than 1/0th the 3:1 VSWR bandwith = approx. 4kHz.  Since one and a half turns of the tuning capacitor tuned the Alex Loop from resonance at 7 through 30MHz (and assuming the frequency response is linear, which it is not) then the required angular resolution is 360 * 1.5 * 4 /(30000-7000) = 0.094 degrees.  Since a typical stepper does 200 steps per revolution, then the required gearing ratio would be approximately 20:1.

With this type of performance the magnetic loop could be very close to resonance at the chosen operating frequency presenting a VSWR approaching 1:1.

73's Gary K6YOA

    Message: 3
    Date: Wed, 2 Dec 2015 20:52:18 -0000
    From: "Dave Lankshear"<dave at lanks.plus.com>
    To:<elecraft at mailman.qth.net>
    Subject: [Elecraft]  [K2] Tuning a magloop with KAT100
    Message-ID: <7B2795A9D35B4139812BF284FD02542B at DaveLLaptop>
    Content-Type: text/plain;	charset="us-ascii"

    Hello, Brian.

      

    I've read your post and the replies and figure I maybe read what you said
    incorrectly, but you did say:

      

    Then use the TUNE button on the K2 to tune the loop to resonance, then
    operate.

      

    You can't use the K2's antenna tuner to tune a loop to resonance.  That's
    not how they work.

      

    The loop is a complete tuned circuit in itself.  There's the loop
    representing inductance and a capacitor in parallel with it.  These are
    resonant at a frequency and the only practical way to move that resonant
    frequency is to adjust the value of the capacitor that's in parallel with
    the loop.

      

    You can't adjust that resonant frequency using an auto ATU at the end of a
    length of coax.  It's rather like using a telephone conversation to make a
    physical change at the other end.  You can't use the phone to make the beds
    back home, when you're away!  The auto ATU will see the coax cable and the
    loop as a lump of L and C and will endeavour to match it to 50 ohms to get a
    1:1 SWR.  The coax forms part of the antenna and is not behaving like a
    transmission line and the loop is not behaving like a resonant circuit - you
    might just as well hook the coax to your automobile's fender and use the
    auto ATU to tune that!

      

    Please forgive my descriptions if I have misunderstood your question, but
    re-reading your quoted statement makes me feel like I'm the only one who has
    understood.

      

    Yes, by all means match the auto ATU into 50 ohms using a dummy load.  That
    way, the PA is looking into a load that matches the impedance of the
    transmission line, although as Don suggested, you don't really need the auto
    ATU (and its losses) as the PA should be reasonably well matched by
    bypassing the auto ATU entirely.  Now, at the loop end, there's a gamma
    match arrangement that ensures the loop, when at resonance, is a decent
    match to the 50 ohm transmission line you're using.  So the rig matches to
    the transmission line which matches to the loop that's been tuned to
    resonance with its inbuilt tuning capacitor.

      

    A loop is only a single turn coil, the resonant frequency of which is varied
    by adjustment of its parallel capacitance.  Because it's small and is a low
    loss inductor operating with a low loss airspaced (or vacuum) capacitor, the
    Q factor, or "goodness" of the single turn coil at resonance is very high.
    This means that a small excursion away from resonance, the loop's Q falls
    very rapidly and renders it pretty useless, thus it is necessary (more so
    when transmitting through the loop) to retune it for frequency shifts of
    more than a few kHz.  That means that the SWR rises rapidly away from
    resonance and the coaxial cable is more involved in becoming part of the
    antenna and less of a transmission line.

      

    The outer surface of a loop (well, outer 6%) needs to be of very low
    resistance in order to maximise Q at resonance.  RF skin effect uses only
    the outer surface of the conductor, thus the larger the surface area of the
    conductor, the more its internal resistance is in parallel and thus reduced,
    so the better performer the loop becomes.

      

    Even a soldered joint on copper piping offers resistance that compromises
    the loop's performance.  Recently a friend gave an old army magnetic loop to
    a group of collectors/militia enthusiasts.  It was in poor condition, but in
    its prime, its surface area must have been a foot across.  This makes a
    mockery of the little bits of aluminium (aluminium) joined together with
    bolts and wing nuts.  Yes, says the vendor, it is broad banded and only
    needs retuning every 100kHz or so.  What he doesn't say is that its
    resistance makesthe Q so appalling that its performance is lousy (where
    lousy is the polite word), but those devices give properly engineered mag
    loops a bad name by tarring all with the same brush.  Also, there are
    proportionally more crappy mag loops out there simply because they are
    cheaper than the "real" thing.

      

    The MFJ 1782/86/88 aren't too bad and are just at the crossing point between
    good and bad, with a bias towards the good, if not too many spiders and
    other insect life are resident under the black covers. These loops have an
    airspaced variable capacitor within the black covers and that capacitor is
    tuned by a small electric motor that's attached to it.  DC power for the
    motor is fed down the coax cable itself, as well as the radio signals.  They
    are not difficult to separate, eliminating the need for a control cable.
    The DC voltage on the coax is reversed in order to make the motor turn in
    the opposite direction.  The more sophisticated MFJ controller has an
    inbuilt cross-needle SWR meter and the name of the game is to get the SWR as
    low as possible on the operating frequency.  Their semi-automatic controller
    drives the motor and detects the lowest SWR point and stops.  Of course, to
    detect the lowest point, it has to begin to increase again, so it always
    stops at a point that's not quite at resonance and this must be fine-tuned
    by the operator, hunting to and fro with up/down press-buttons.  It takes
    more effort to describe than to do in real life Hi!

      

    One last caution.  Circulating currents are very high in a transmitting mag
    loop and very high voltages (thousands of 'em) are also present when at
    resonance, so ensure that the loop can't be touched by anything that matters
    to you when it's transmitting, even at a few watts.  Give it a LOT of
    respect, indeed, powerline respect.

      

    I hope my efforts haven't missed the mark by a mile, Brian and that instead,
    the wear and tear on my keyboard has been of some small use to you.

      

    73 and early Season's Greetings.  Dave G3TJP