ad: Momobeam-1

Loop-On-Ground Antenna

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by AI5DH, Aug 1, 2018.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: HRDLLC-2
ad: L-MFJ
ad: MessiPaoloni-1
ad: abrind-2
ad: L-Geochron
ad: Left-2
  1. N3DT

    N3DT Ham Member QRZ Page

    I've got a dog fence underground wire (surrounds about 4-5 acres) that I've been thinking about pressing into service as a loop underground (LUG). I actually have some RG6 running up to where the wire connects to the dog transmitter, but I haven't done anything. I did get some binocular cores to make a 9:1 transformer, but just haven't gotten around to it yet. But I'm wondering since I live in the boonies and the noise floor is so low here I may not get anything from it. I tried a small loop once and that was a total failure compared to my 65' dipoles and H DoubleBays. Noise floor level is not an issue for me.
  2. AI5DH

    AI5DH Ham Member QRZ Page

    OK then IMO you are not a candidate for LoG as you have plenty of room and low noise environment. Having said that real easy to test. Take a radio out and hook it up crudely and see what happens.
  3. AI5DH

    AI5DH Ham Member QRZ Page

    Not sure I am the right person to answer all your questions or not. I am an electrical engineer, but EE's are a bit like doctors in that they specialize. You would not go to a Proctologist for a brain tumor. Well come to think of it, I know a few here that should. But I will give it a try. Start with this document as it entails exactly what you are trying to do.

    I am a simple man and I use a KISS design practices (keep it simple stupid). The first two pieces of data you need to know:
    1. Lowest operating frequency in KHz. (500 KHz in this example)
    2. Impedance of the antenna. Will be used to calculate XL of the coil impedance (450 Ohms in this example)

    These two bits of data determine everything. Get it right and the rest pretty much takes care of itself. First step is to find XL where XL is the impedance of the coil winding in Ohms.

    XL = minimum of 4 x Antenna Impedance.

    It can be 5 or 10 times higher, but 4 is minimum and determined by lowest operation frequency. Get that right and higher frequencies take care of themselves. So XL = 4 x 450 Ohms = 1800 Ohms = XL. Easy peasy so far.

    Next is to find the number of turns it will take, and to do that we must first solve for L or the inductance coil.

    L = XL/6.28f

    Where L= Inductance in millihenries XL=Reactance in ohms, f=Lowest frequency of operation in kHz. So 1800 / 6.28 x 500 = 1800 / 3140 = .573 mh. That is a huge inductance FWIW. Still easy peasy

    OK now we are ready to determine how many windings it is going to take, and to do that requires a bit of homework. This is where selecting the right Core Mix and physical size of the Toroid doughnut comes into play. Let’s make it a bit easier on Core Mix. If you live in the world of 2 MHz to 30 MHZ it is as easy as Mix 43, and sometimes 73. However I am looking at 500 KHz and mix 75 is the best choice for 200 KHZ to 15 MHz. OK that takes care of Mix, I used 75. Now the hard part; what size doughnut or binocular is required?

    This gets into your question; can I use a small used Life Saver Candy sized doughnut since there is no real power involved in RX only antenna. Answer is NO in this example. You are right in that when TX power is involved, physical size and mass are major contributors to handle the heat involved. Requires thermal mass to handle power and heat. But we need a lot Inductance and that is also directly proportional to mass and physical size. Mix 75 has the highest Permeability Magnetism of all mixes, but lower frequency application restricted. The Constant Value we are looking for is AL in our doughnuts. So have a look at Fair-Rite Mix 75 Toroids and spot AL column. You will see later from the math I will need something on the order of 5000 and greater for AL. I selected Fair-Rite 5975007601. Not huge at 15 grams weight and .88-inch diameter. FWIW now go look at other core mix AL values and you will quickly understand why 75 is the only choice in this example. I am using AL =6100

    OK to find windings required:

    N = 1000 x Square Root L/AL

    Where number of turns required (N) is equal to 1000 times the square root of the Inductance (L) divided by the constant AL. So here we go:

    N = 1000 Square Root.573 / 6100 = 1000 Square Root.000093934 = 1000 x .009691977 = 9.69 turns minimum. OK 9.69 is not a whole number, so at a minimum need to bump that up to 10 turns. However in this example 10 turns does not work great because we want a 9:1 impedance ratio for a 50 or 75 ohm coax, and that works out to a Turn Ratio of 3:1 (Square Root of Impdeance Ratio). So if we bump up to 12 Turns (greater than 4x requirement is a good thing) works out perfectly with a secondary of 4 Turns.

    As for a sanity check to determine if I made the right core and mix choice is to look at how many turns we used on the Low Side of the Transformer. If we used 1 to 4 turns we got it right with 1 turn being perfect. OK ready to wire it up. Being a RX only we could use 30 AWG, but this core can accommodate a larger wire which helps make the Impedance a little higher which is a good thing because we want a minimum of 4X or higher being better. Old schooled would use magnet wire. Smart money today uses solid copper, silver plated, PTFE Insulation (Teflon) Hook-Up or Wire Wrap because it is superior in every way to magnet wire or any others. Both use solid copper. However RF runs on the surface of conductors called Skin Effect and nothing has lower resistance than silver. Teflon insulation has superior dielectric/insulation strength, tough as nails abrasion resistance, an operating temperature up to 450 degree F, comes in colors, and a friction coefficient slicker than snot on a doorknob while threading those windings through those small binocular core holes.

    Hope that helps.
    Last edited: Aug 3, 2018
    N5YPJ, N0TZU, K0OKS and 1 other person like this.
  4. N0TZU

    N0TZU Platinum Subscriber Platinum Subscriber QRZ Page

    I've been playing around with this LoG on 80m (not much happening on 160m) for a day since seeing Derek's post. I have about a 15 x 20 area and using an isolation transformer (9:2 windings on type 77 core that I had around).

    Other than one instance out of many, so far I'm not getting better perceived SNR than my vertical in A/B comparisons. I try to use the preamp and AF gain to get the same or close "by ear" signal level in each case. RF gain is maximum for the comparison. Before anyone jumps at the core type and number of turns, I also tried it without the transformer with similar results.

    I also tried some BCB listening, and NDB listening below 500 kHz with the same results, although there might have been improvement in lightning static crash noise but it was hard to be sure. That would be highly dependent on pattern and distance to the storm so one listening session isn't definitive.

    I'm suspecting that it's just the low noise floor here making improvement difficult, at least for the directions and distances of stations I encountered. Also, the LoG antenna is pretty close to the house since I didn't go to the trouble of laying out a long feedline, and that could be causing some pattern issues with it since the house is stucco.

    I also have a low dipole that I can sometimes use to get better SNR than the vertical depending on station location and type of noise, but couldn't compare to it because I used it's feedline for the LoG. I'll switch things around later and compare.
  5. K0OKS

    K0OKS Premium Subscriber QRZ Page

    KF5LJW, thank you for taking the time to post that lengthy response. Your post, and the linked document is exactly what I was looking for to fill in some gaps in my understanding of transformer core selection. Thanks!
    N0TZU likes this.
  6. K0OKS

    K0OKS Premium Subscriber QRZ Page

    Here are my preliminary results...

    I am using what I describe above 15' square loop, 9:3 turns on Mix 31 2.4" OD torroid, SO-239 and 50Ohm LMR-400. This goes into my MFJ 1040C preselector (which also has a basic FET based preamp). This goes into the MFJ1026 Noise Canceller/Signal Enhancer.

    I ran a Frequency Response Analysis on my transformer before I installed it. This is a pretty poor version provided by the latest Keysight scopes. It is best used in direct comparisons, as it is quite limited. Anyway, the peak was closer to 10MHz, rapid falloff up above 13 or so, slower falloff towards 2MHz.

    In actual on-air listening I am finding that I can garner some benefit by phasing the LOG with my main Windom based antenna using the MFJ 1026. Direct listening to the LOG does make loud stations quieter (great S/N ratio), but does not seem to help much in bringing out weak stations, even with varying amounts of preamplification (in the shack, not at the antenna). Phasing the LOG with the main antenna does let me reduce the noise a little bit and it seems to pop the weak signals maybe 1-3 dB, not a night and day difference, but can be useful.

    The results are much more pronounced on 40m, but a tiny bit effective on 80m and mildly effective on 20m. I think I definitely more performance on 80m. So I will try a mix 73 setup.

    I have a unique PITA issue in that my location for the LOG crosses over my burred UVerse (ATT internet) cable. This is bad new because UVerse creates a LOT of hash right around 3.8MHz, up to about 4MHz or beyond. I can clearly see this enhanced pickup. I may try to put some chokes out at the pole end. They just burred the cable, and they did install a box at the pole. So there is some room to get at least some choking in there. Probably not enough cable for the number of turns required, but it might help (even if ATT might not like it). I also may try to move the LOG about 10-20ft over, which would likely help a lot.

    So I shall experiment with different transformers, and perhaps move the anttena over some, and maybe choke the stupid UVerse line. (Anyone else using UVerse... try putting a mix 31 clamp on (big one) with as many turns as you can get through it on your line coming into your house (I have two, each one is about 13 turns I think). It will make a BIG difference on that noise that starts right at 3.8MHz.)
    KA0HCP likes this.
  7. AI5DH

    AI5DH Ham Member QRZ Page

    THX for trying and sharing, I appreciate it. Like I stated the LoG is a Niche Application antenna for the lower bands for us poor SOB's (strike 1) , plus restricted size lots or HOA (strike 2). with high noise environments (strike 3) that have trouble with RX on 160/80 and even lower frequencies. 3 strikes and you are SOL. Being my thread and biased optimist I can ascertain from what you have said is the LoG works. From you experiment and observations you stated you did not experience any gain which I do not have any doubts, but essentially you are saying works as good or almost as good as your elevated dipoles performed. Would that be a fair conclusion?
  8. AI5DH

    AI5DH Ham Member QRZ Page

    KOOKS and NOTZU thank you both for sharing. Just a couple of passing thoughts since you went this far and might help in my experience with it. If you wish to take it to the next step, optimize the antenna. There are a few points to follow to get the most out of the LoG. You can get the details from KK5JY BOG but the main points are:

    1. Loop is Bi-Directional like a Dipole.
    2. Symmetry is important. Calling it a loop is a bit misleading, SQUARE is most accurate. 15 x 15 feet is recommended or 60 feet of wire.
    3. Where Loop is Terminated is critical

    Those are the main points to optimize the antenna in my experience, all 7 days now. As for a cores, I found an available part perfect for the job an Amidon FT-87A-75. I have a couple of extras I will let go for $5 each. They cost me $4 each because you must buy 10 of them and shipping is $14.

    Another thing I do that is not common is I am running Balanced, or I should say Floating Antenna. On my SDR RX I use the Hi-Z 1000 Ohm Balance Port. At no place or time is the antenna referenced or bonded to ground or dirt. I do use coax, but I use Isolation Transformers on both ends of the coax. Point is the COAX is not grounded, it is Floating in a Balanced mode. Not sure if that helps performance or not, just bringing up the point if you use coax direct to a radio, you grounded the coax. Might try a 1:1 isolation Transformer at the radio to see if it helps is common mode issues pop up.
  9. N0TZU

    N0TZU Platinum Subscriber Platinum Subscriber QRZ Page

    I don’t doubt that it works, and I did have one instance where I did notice some SNR improvement. I agree that it probably really shines if the local noise floor is high which isn’t my situation, so that’s likely the main reason I didn’t experience an improvement in SNR relative to my vertical in almost all cases.

    Again, though, it was limited testing and I didn’t put any effort into making sure the signals were coming from various compass directions, so the directivity aspect wasn’t explored thoroughly. The bottom line is that for my situation there didn’t seem to be a dramatic SNR improvement with the LoG. (Also, the gain was low enough with the LoG that some signals couldn’t be heard at all on it, even with the receiver preamp engaged).

    I haven’t yet tested it against the low dipole because I used part of its feed line for the LoG. I will rearrange and check that. As I said sometimes I can get better SNR for a station on the dipole because the signal vs noise is favored more by the dipole pattern either in azimuth or elevation or both, compared to the pattern of the vertical.

    (One very clear difference I find with a ground mounted vertical compared to a horizontal dipole (or loop) is its sensitivity to local power line noise, at least when the noise current travels down the pole ground wire, which is of course vertical and close to the ground where the take off angle is low)
  10. K0OKS

    K0OKS Premium Subscriber QRZ Page


    I moved the LOG about 15ft to one side, so it is no longer on top of my buried UVERSE cable. WOW! YUGE difference! This got rid of all the noise on 80m (which was defeating the purpose of a low noise antenna). It also reduced the noise on 40m quite a bit. Could just be the new location is better overall. It is 15 ft farther away from my computers as well. So now this antenna is working very well indeed. I am pulling signals out on 80m (75m SSB) with JUST the LOG. These signals are difficult or impossible to copy on my Windom variant due to all the noise. So it is indeed working to increase S/N ratio! I mean quite a bit, easily 3dB, possible more like 6dB (an S Unit).

    Interestingly enough this version is less symmetrical than my initial attempt (I did not measure it this time, I just moved the wire and stakes over... last time I measure it all and used as T-Square). I believe this one is more squished in the "Y" dimension... in other words the sides are farther apart from each other than the feedpoint is from the opposite end.

    I still plan on experimenting with the transformer.

    KF5LJW, interesting idea about the isolation transformers. Might be worth trying for the heck of it. Did you make those out of the same cores, or are those little storebought isolation transformers?

    Anyway, after my moved I am very pleased with this antenna's performance.
    KA0HCP likes this.

Share This Page

ad: cq2k-1