What is my sky loop antenna doing?

[ChrisSmolinski]

I was curious about how my sky loop antenna actually behaves, resonant frequency wise. I've always assumed the antenna is about 670 ft in total perimeter. I say about because it has been both extended (from the original 635 ft) as well as repaired after several storms over the years. It is entirely possible it is a different length, possibly shorter.  I took some quick and dirty measurements and come up with these measurements for the four sides of the loop:
140
160
165
185

So a total of 650 ft. I paced it vs dragging out a tape measure, so assume a +/- 10% error easily. 

In theory a loop antenna is resonant at one wavelength, and at odd harmonics. If we split the difference and say the loop is 660 ft, that would be 1477 kHz. My loop is not circular, but rather more of a deformed quadrilateral. Throw in the fact that the height varies above the ground, it is run through trees, etc. and the actual resonant frequency can be very different.

Using a noise bridge, I tried to identify all the resonance frequencies I could find (those where the reactive component of the impedance was zero). Here is what I found, along with the resistive component at that frequency, as well as the number of kHz this resonance was above the previous:

kHz   Real Ω   kHz above previous resonance
1950  20 ohms  -
2630  25 ohms  680
3740 75 ohms  1110
5850 30 ohms   2110
6490 40 ohms   640
10260 40 ohms  3770
14350 40 ohms  4090
18430 40 ohms  4080
22480 40 ohms  4050
26620 40 ohms  4140
30770 40 ohms  4150


The antenna is fed with a balun (I *think* 4:1, did I mention the antenna has been up for some time?)

For the higher frequencies there's a nice roughly 4100 kHz spacing between resonant frequencies, which seems to imply the fundamental resonant frequency is 2050 kHz. The first resonance measured is 1950 kHz, so that seems to fit.

There's several rather strange resonances in the list above, namely 2630 and 6490 kHz. I suppose this could be due to some interaction between this antenna and other antennas (there's an approximately 200 ft beverage that runs down the center of part of it, at a much lower height, as well as a tilted folded dipole for the 48m band near one side, but also inside).

I had two primary reasons for looking at the antenna. First, how does it perform on the 43 meter pirate band, my main area of interest. While there's no resonance in this band, from experience I find it works well there. I also compare it to my 48 meter band tilted folded dipole, and it is similar, sometimes one antenna wins vs the other, depending on the station.

The second reason is for performance around the 4096 kHz pirate beacon band. I used to have regular reception of these beacons every morning. Then, a few years ago, they almost completely vanished. It could be propagation, but it seems odd that I would go from daily reception to almost never hearing them. That made me wonder whether something with the antenna changed, perhaps from one of the repair jobs. I do notice the resonance around 3740, and if the antenna was somewhat shorter, that would be moved to around 4096 kHz. But the antenna performs well over most of HF (actually it even works into the VHF band, once I got the downconverter for my netSDR I started to explore VHF and even UHF with it.), so I am not sure how important the resonance is.

The antenna does work very well on the upper half of the MW band, as you'd probably expect.

I have tried modeling the antenna with NEC, but, well, you know how that goes. Nothing seems to match reality with an antenna like this, and a lot of unknowns and guesses. Maybe I can use NEC to predict lottery numbers instead 

I'm looking for feedback from anyone else who might have had experience with a similar antenna.

[Josh]

I've had several loops of the kind, none as large a capture area as yours but they were several hundred feet at least and up 14 to 30ft or so. Outstanding hf antennas, actually they work great at mw and vlf too. I just fed them with coax and some ferrites at the feed point. I never bothered getting sciency and determining where they were resonant as I always used a tuner.
They were such that a TS850 and IC756p (and several others) could tune every band easily with their internal tuners. A friend has one that's about 500ft atop phone poles, it presents around s9 band noise on 20m in a ssb bandwidth, and he lives in an rf quiet place.

[Terry]

No experience with those type antennas but I do have some lottery experience. NEC couldn't do any worse than the quick pick on the Lotto machine.

Interesting post though. I may try a loop but it will have to be smaller since I don't have much room.

[Fred Smith]

Have you checked to see if the wire is broken along the loop. It might be acting as a dipole antenna instead. To me, the numbers suggest that.

[ChrisSmolinski]

Have you checked to see if the wire is broken along the loop. It might be acting as a dipole antenna instead. To me, the numbers suggest that.

Yes, I actually did that yesterday. From experience when the loop has opened (such as from a tree falling on it) and it turns into a large dipole, the noise level goes up tremendously, which I am not seeing now.


I just realized an added complication - I am measuring here in the shack, after 100 ft of RG-6 coax, which is transforming the actual antenna impedance.

[RobRich]

What height? Type of soil? In particular, I am referencing angle of radiation and ground losses.

At low height, your loop very well could be NVIS at low frequencies. YMMV for DX, especially if propagation is lacking.

You loop also is developing more peaks and nulls in its horizontal pattern as frequency increases. Again, YMMV depending upon if the desired received signal is in a peak or null.

I do not bother with impedance matching for my ground-level 148' "shielded" loop. I even pulled the balun a long time ago and instead loaded up the feedline with snap-on ferrites plus an 1:1 CMC choke near the receiver - yeah, I was being lazy about the choke placement.

[ChrisSmolinski]

The height varies dramatically around the loop, based on what trees I could use     Mostly in the 25 to 45 foot range I would say.

I agree that there are likely strong peaks and nulls around the horizontal pattern, and these probably move around a little as the wire changes position with the trees for various reasons.

I'm generally happy with the performance, I pick up a lot of pirates which is my main interest, and it works well over most of HF.   There are a few pirates that are notoriously difficult for me to hear well, such as Rave On Radio, which could possibly lie in one of those nulls.

My investigation was driven mostly by the disappearance of the 4 MHz pirate beacons about two years ago.

[Fred Smith]

Has anyone else heard the pirates?

[ChrisSmolinski]

Has anyone else heard the pirates?

Yes. They are being continuously logged by others on the HF Beacons forum here. I have heard them one or two times since they mostly vanished, but not on the regular basis as before.

[RobRich]

Mostly in the 25 to 45 foot range I would say.

Then the angle of radiation/intercept should be mostly NVIS at 4MHz. Propagation is going to play a big factor in long-haul DX intercepts on a NVIS antenna, and activity within the current solar cycle had a secondary peak in 2014, with a slide afterwards.

Assuming the beacons are distant DX, there is a chance you might fair better with a ~58' vertical (wire in tree?) fed against a few ground radials, assuming it is not overloaded with local noise anyway.

[ChrisSmolinski]

Yes, the beacons are DX, most are in the SW USA in the desert. It is indeed possible that I was hearing them for a long period due to solar / propagation conditions which have since changed.

I have been considering putting up a folded dipole cut for 4.1 MHz. The orientation for best reception to that area would also work well for stations from Africa, and it would probably work fairly well on the 60 and 90 meter bands. Although CODAR has really trashed 60 meters.

[RobRich]

Thinking along the lines of a folded dipole, a T2FD might be worth a look, too. Both designs need a balun anyway if fed with coax. A receive-only T2FD should suffice with a comparatively small non-inductive resistor versus a costly larger resistor for transmitting. You are giving up gain for bandwidth, but a simple FET preamp circuit should offset losses if needed.

Regardless of the design, height is still the big issue for angle of radiation/intercept and broadside directivity of a DX dipole. A ground-mounted vertical or an inverted L might offer an alternative DX solution if getting a dipole high enough for low-frequency DX proves to be an issue.

[ChrisSmolinski]

I had a T2FD for many years in fact. It eventually got destroyed from too many storms. IIRC it was 132 ft long, and worked well down to the 120m band. Might be an interesting idea to build another, vs a folded dipole. At least I won't need to special order ladder line for it. My old one used wooden dowels as spacers, I think I might go with PVC this time. The wood is what failed first.

Thanks for reminding me of the T2FD again, I'll have to seriously consider it. It will probably be late summer or early fall before I install anything new, so plenty of time to think about the options.