ad: vanity

Impedance Matching of Untuned Loop RX Only Antennas

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KK4NSF, Nov 22, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: abrind-2
ad: Left-2
ad: Subscribe
ad: L-MFJ
  1. KK4NSF

    KK4NSF Ham Member QRZ Page

    Hello All, Continuing my work on untuned and/or untuned-active small loop antennas for RX only, I've been calculating Impedance Matching Transformers based on the work of VK3HK, ( Stephen Bushell) , KK5YJ (Matt Roberts), and W8JI (Charles Rauch, Jr). Their work appears to be spot-on, and is greatly appreciated.

    Using VK3HK's formulas, and KK5JY's sketch I built a spreadsheet for calculation the required turns to match a wide range of antennas to a 50 ohm coaxial cable. I used a conventional transformer in my calcs.... not because I think it's better, but because it was the easiest to alter in order to change the impedance for my experiment. Here is a sketch of the transformer... and the spreadsheet is also attached at the bottom. (It's a pdf, but an Excel file is also available if anyone wants a copy)

    transformer-schematic.png
    (yes Matt.... I swiped your photo, sorry....but I do give you all the credit ! :D)

    Using the VK3HK formulas, you can easily figure out that the impedance of KK5JYs loop is ~10 ohms. To confirm this, I built a copy of the antenna using a ferrite toroid... and tested it using my SDR Spectrum Analyzer, on MW, 40m and 20m, using the floor noise as a reference. Without the transformer, both the floor noise and the signals are noticeably weaker. With the transformer, both the noise and the signal increased dramatically, so it looks to me like the transformer is working.

    As designed, the antenna works very well on receive, and has the broadband characteristics I'm looking for. It appears to work from the MW Broadcast Band, up past 20m. I didn't look any higher.... so it may be great, or may be a dog.

    To test this further, I rewired the test antenna using a 1:1 turns ratio (50 ohms to 50 ohms), and the signal dropped noticeably on the waterfall, but the noise floor did not. A third test at a 3:1 turns ratio (450 ohms to 50 ohms) produced a drop in both signal strength and noise floor. SO... I'm confident that the KK5JY / W8JI transformer does produce a good match to the 50 ohm coax, and VK3HK's formula produces an accurate result. It also appears to confirm that the impedance of the KK5JY loop by itself is in the neighborhood of 10 ohms at 7 mhz.

    Now..... my next step is to build a 5'ft square version, and do the tests again to see if the transformer functions the same with the larger loop, OR if the 5ft untuned loop's impedance will be higher, and thus require more secondary turns to produce the same match. Unfortunately, I've not been able to find the formula to calculate the impedance of an untuned loop..... So I'm just going to figure it out using an antenna analyzer, and / or experimentally.

    More results, successes, failures to follow....
    Dave
    KK4NSF

    (constructive and/or helpful comments would be appreciated)
     

    Attached Files:

    Last edited: Nov 22, 2019
    N0TZU likes this.
  2. KI4ZUQ

    KI4ZUQ Ham Member QRZ Page

    2t:5-6t = 2.5 to 3 to one. But a 3:1 transformer had less noise and signal? Please explain...
     
    KK4NSF likes this.
  3. VE7BPB

    VE7BPB Ham Member QRZ Page

    A quick way to play with the turns on a larger ferrite like a 240-? is to use standard test bench clip leads for the winding. Makes it very fast to change the number of turns. I use a chassis mount style BNC connector with a couple of short leads soldered on for the clip lead to connect to.

    regards, Roy
     
    KK4NSF and AK5B like this.
  4. KK4NSF

    KK4NSF Ham Member QRZ Page

    Good question.... and one that tripped me up at first, too.

    Impedance Ratios are not a simple "number of turns" sort of thing like a power transformer, but are the square root of (Zp/Zs) thus a 3:1 Turns ratio results in a 450ohms : 50 ohms or a 9:1 Impedance Ratio. The 9:1 impedance ratio produces a mismatch.... which decreases the signal.

    A 2 to 5 Turns ratio results in an almost 5: 1 Impedance ratio. Or 10 ohms to 50 ohms...
     
    Last edited: Nov 22, 2019
  5. KK4NSF

    KK4NSF Ham Member QRZ Page

    That would work. For my little experiment, since the antenna is outside, I simply disconnected and rewound the transformer to a predetermined ratio.
     
  6. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    Another option is to use a very high impedance preamp at the feed point and make this an active antenna. I use the AD8067 op amp for all my active antennas now. The chip has basically infinite input impedance, and is normally run with about 10:1 voltage gain. The nice thing about this scheme is that impedance matching to the transmission line is just about irrelevant. I hope you get a chance to check out my ARRL book, Receiving Antennas for the Radio Amateur. I talk a lot about active antennas of all kinds in there.
     
    N0TZU and KK4NSF like this.
  7. KK4NSF

    KK4NSF Ham Member QRZ Page

    Eric - I've read that book.... and admire it greatly. It's one of the things that inspired me to look at receive-only antennas in the first place.

    You are absolutely correct about making this an active antenna.... and that is part of where I'm going with this project. My end goal is to produce eight identical low-cost receive antennas (~3 to 6ft diameter) , connect them to an array, and then use a preamp at the array feed point. The reason I want as close a match as possible to 50 ohms is so I can use it as a passive antenna (and bypass the amp) if needed to avoid overloading the receiver front-end. A good match makes that a simple task.

    Dave
    KK4NSF
     
  8. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    Understood.
    In that case you MAY want to revisit the broadband transmission line idea. Using "normal" transformers will yield unpredictable phase shifts...and for the array you're building, you don't want that! So I'd stick with a "standard" design, even if it's not absolutely optimum. For phased arrays, consistency is more important.
     
    KK4NSF likes this.
  9. KK4NSF

    KK4NSF Ham Member QRZ Page

    Agreed. I'll look at that. Thanks
     
  10. K4SAV

    K4SAV Ham Member QRZ Page

    The impedance of the 15 ft on a side LOG isn't close to 10 ohms. Without the transformer, the real part of the impedance varies from about 12 to about 500 ohms on the bands 160, 80, 40, 30, and 20 meters. The composite impedance varies from about 300 to 1500 ohms on those bands. That's why a 9 to 1 transformer gives a pretty good SWR on all but 160 meters when connected to a 75 ohm feedline. The actual SWR isn't very important for a single element receiving antenna. It becomes important if you try to phase them. The antenna has a parallel resonance close to 5 MHz, where the impedance is very high.

    Including the transformer and 150 ft of RG-6 the measured SWR was 6 to 1 on 160 meters and below 3 to 1 on all frequencies from 3.5 to 17 MHz.

    Those numbers came from a measurement I made at my house. Measurements at your house will probably be different. Similar measurements made on antennas on the ground show a lot of variation with ground moisture content and the exact distance above the dirt. Fractions of an inch in height changes the impedance significantly. That will make phasing antennas very difficult. You will have to do the phasing experimentally. You will not be able to use NEC to predict the proper phasing. NEC doesn't come close to the the measured impedance and phase angle. You will likely need a variable phasing network and adjust it each time you use the antenna.

    There are much better alternatives but I encourage people to evaluate antennas like this themselves.

    Jerry, K4SAV
    LOG Z.jpg
     
    Last edited: Nov 23, 2019
    N0TZU and KK4NSF like this.

Share This Page