90-Degree Phasing of Two Mag-Slots for Endfire

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KY8D, May 23, 2018.

ad: L-HROutlet
ad: l-rl
ad: abrind-2
ad: l-BCInc
ad: l-gcopper
ad: QSOToday-1
ad: L-MFJ
ad: Left-3
ad: Left-2
  1. KY8D

    KY8D Subscriber QRZ Page

    I have been playing with EZNEC, iterating my way through a dozen different variations of phased, triple magnetic slot antennas. When I have a pair working perfectly, the pair are double the impedance of either one alone. Such that if I build the perfect 100 Ohm endfire array, then delete the phasing harness and 2nd antenna, the single mag-slot remaining will be 50 Ohms. Like that whether the pair together are 200, 100, or 50 Ohms. The orphaned half of an endfire pair is always half the impedance of the former pair ... and with lowest SWR just slightly higher than when paired for endfire.
    To feed a pair of 50 Ohm mag-slots for endfire, EZNEC insists on two legs of 100 Ohm feedline, one 1/4 and the other 1/2 lambda (or 3/4 and 1 lambda). And to hook into those with 50 Ohm coax, EZNEC wants a pair of 1:2 baluns.
    Another thing I noticed is that the lower the impedance, the sharper is the SWR notch. Such that a single 25 Ohm mag-slot will cover only half of the 10m band. A 50 Ohm mag-slot covers all of 10m, but only barely. A 100 Ohm mag-slot is nice and flat (1.2:1 worst case) across the whole band. When paired as endfire, the impedance doubles and also the SWR flattens out just like for a single of that same impedance. I modeled a pair of 100 Ohm mag-slots together for 200 Ohms endfire, and SWR was as flat as can be.
    So my question for the experts is this: am I trading away anything by going up in impedance so as to have a flatter SWR? A nagging voice in the back of my mind keeps hinting that since I'm lowering Q, I might be sacrificing the advantage of quietness. Will a 200 Ohm mag-slot be less quiet than a 100 Ohm? I really want to be keeping the quiet for clean reception. All the above (the sum of my knowledge of mag-slot antennas) was gained by trial and error, playing with values in EZNEC, observing the pattern, and noting results. I don't have any theory at all. And EZNEC tells me nothing about if the antenna will be either quiet or noisy. That nagging voice comes from reading lots of articles also on magnetic loop antennas, which everyone says are ultra quiet ... and laud their ultra-high Q. If my SWR is getting flatter, surely I am lowering Q. But ... how to be sure. I want to be sure. I'll be investing some very expensive wire.
    Gan Starling, KY8D
  2. AK5B

    AK5B Ham Member QRZ Page

    Whenever one lowers Q it increases the bandwidth and lowers efficiency; time to pick your priorities. I wouldn't obsess over minor SWR changes, either.
  3. AI3V

    AI3V Ham Member QRZ Page

    Build each model, measure the results, and get back to us with the accuracy of your model.


    P.S. for a start, the impedance of an antenna will have exactly zero effect on it's ability to magically separate signal from noise.
    AK5B likes this.
  4. K4SAV

    K4SAV Ham Member QRZ Page

    I'm not sure who invented the term mag-slot. Sounds like something recent. However the antenna has been around for a very long time. A simple one wavelength vertical loop fed on a side with dimensions that cause the feedpoint impedance to be something close to 50 ohms, or whatever your feedline impedance happens to be.

    Yes you can phase them. The biggest problem is that the beamwidth is very narrow, even for a two element version. Unless you have only one place to point it, it would be much better to make it rotatable. That should be doable for 10 meters, however with the length of the boom you will need, you could put up a multi-element Yagi that has a lot more gain. Even a three element Yagi would kill the loops, and be a much smaller antenna. Of course you could put up 4 of the fixed loops and switch between them to cover 360 degrees with only about 3 dB loss due to pointing error.

    The best use for an antenna like this is probably 80 meters, where it is difficult to get low angle gain because of height limitations for horizontal antennas and the fact that most people don't like the problems associated with radials for verticals, plus the fact that you don't need a matching network.

    An antenna's "quietness" is a function of its pattern. Use the RDF number for a reference of how quiet it will be. EZNEC will calculate that for you automatically. Do a 3D plot and take the gain at 20 degrees elevation and subtract that from the average gain number displayed at the bottom of the main window. If you are working with something other than low band antennas you may want to use a reference other than 20 degrees. For more information see

    Jerry, K4SAV

    My 1975 GTO had mag-slots.
    KY8D and AK5B like this.
  5. KL7AJ

    KL7AJ Ham Member QRZ Page

    This statement needs to be qualified. Fatter antenna elements generally lower the Q while also increasing the efficiency. However, your assertion is correct for sharply tuned antennas like the small transmitting loop....or Texas Bugcatcher.
    KY8D and AK5B like this.
  6. KY8D

    KY8D Subscriber QRZ Page

    The first is simply not going to happen. At least not models for 30m. Not with the wire I plan to use, and which is already purchased: 10 GA, 37-strand hardened aluminum, counter-wound. Three lambda worth of that for each antenna of the pair. I've done a skin-effect calculation and that gives me a 1.5 advantage over DX Engineering's "premium" antenna wire of 14 GA, 19-strand copper, which I've been using for wire antennas up until now. Most annoying is the effect of how very strongly 100 feet worth of the simple-twist wire tries its very best to untwist when under even just five pounds of tension. This more expensive wire will not stretch, will hold its shape, and will exhibit far less catenary sag for the same span. But it's too expensive to play around with making six different versions. Maybe someday, if ever I work my way up to the 6m band. ;-)
  7. KY8D

    KY8D Subscriber QRZ Page

    An old antenna design yes. I first read of such in the 2nd and 4th Antenna Compendiums. And indeed, one of those was for 80m. A so-called "double magnetic slot" it was termed. Based on the idea of an actual slot antenna (cut into sheet metal) but reduced to only the skeleton of the slot itself. That one, however, was folded back upon itself. I've modeled several versions of those too in EZNEC; and the numbers do not at all impress. That particular "double" is really just the same identical "slot" executed with double the wire. Twice the Ohmic resistance to no good effect. As elsewhere described, when not thus folded back on itself, the gain is much better. But of course that takes up twice the space. Then is when you have a true double. Extend it to triple and the gain is better still: enough so as to be still worth the effort. Quadruple, however, is past the point of diminished returns. So I figured to build a triple.

    I have a Goldilocks situation on my lot. Too many trees crowding in for a tower supporting any decent length of boom. Not unless I want to go above 75 feet. The house sits atop a bit of slope on two of four sides. The nearest ground that's nice and flat for digging in a tower base is too close to my property edge. Trees elsewhere stand in the way of guys. They're nice trees. I like them. Not going to cut a bunch down. But I have a couple of avenues though between all the trees to aim a long broadside array: bobtail curtain or magnetic slot, with only minimal trimming of limbs. And one of those is aimed nicely for South America and Alaska, and long enough for a triple mag-slot on 30m ... which is where I spend most of my time in pursuit of CW rag chews. A shorter avenue between trees will give me room for a similar pair of endfire mag-slots for 17 meters, pointed at Europe and the South Pacific. Figured to maybe do that one up some later year.

    So the 3-element Yagi-Uda array, while commendable in every respect, will not likely happen for 30m. Not on my lot, and not for wanting to be so high up. I'm wanting some kind of loop. In the past I've had huge, random-length horizontal loops, even huger random-length end-feds; hundreds of meters long in both cases. My aim here is to have at least a single antenna that's band-specific. As the sun-spots return, each year I'm hoping to put up something for the higher WARC bands. Mag-slots I can hoist between masts, and swap out year to year, rolling up the prior onto cardboard tubes like spools for easy storage. One year I'll finish the small magnetic loop already started.

    Thank you kindly for your suggestion on how to predict quietness inside of EZNEC. I shall presently do that. As for 20 degrees, however. All of the 30m mag-slots that I've modeled so far, exhibit peak gains between 15 and 17 degrees, which I figure will be very nice. All have been modeled above average ground with their top wires all at 12m height. Some gain is lost in each step progress from 25 Ohms Z through 50, 100, 200, 300, 450 and 600 Ohms Z. Which is as makes since, since the tripple "slot" apertures become more and more square, such that the bottom wire gets closer and closer to the ground.

    In fact, my original intent was for a 60m mag-slot, for which I have room for a double-aperture mag-slot in that same location. But with the power restriction there being effective radiated power rather than peak envelope, it soon became apparent that my increased gain would require I reduce power ... which seemed partly missing the point. So I went up one WARC band in my ambition.
    Last edited: May 23, 2018
  8. KY8D

    KY8D Subscriber QRZ Page

    This much is true, about 20 to 30-ish degrees for the -3dB points in most cases for 1/4 lambda separation fed at 90 degrees. On the great circle map, this still gives me nearly all of South America, Western Canada, Alaska and (just maybe) as far as Japan. So long as it's not my only antenna, I'm fine with that.

    I'll be stringing the pair between one sturdy tree and three portable masts. After a few years, I can re-arrange things maybe.
    Last edited: May 24, 2018

Share This Page

ad: elecraft