Gear: Antenna experimentations

I still owe a long post on the details of building wood-frame tunes loops, but I have to put together all of the various photos, instructions, etc. One of the problems with these loops is that their figure-8 reception pattern means that if there is a strong station 180-degrees from the weaker station you want to hear, you are stuck. However, there are various ways to get a different reception pattern. The down side is that many of those antennas require a lot of real estate and need to be supported in special ways.

However, I finally came across Bruce Conti’s Super Loop design. This is similar to other specialized “large” loop antennas, but the loop shape is simple (a rectangle), and for those of us who live in rocky areas(**) with poor ground conductivity, it does not require a specific ground. The two bottom corners of the loop are critical; the corner pointing towards the direction you want to null is terminated with approximately 1000 ohms of resistance, and the other end is linked to the coaxial radio feed line via a high-turn transformer to match the high impedance of the loop to the 75-ohm or 50-ohm impedance of the coax leading to the radio. In my case, “the radio” is an inexpensive RTL-SDR dongle plus a Ham It Up upconverter.

I have the dual problem of barely having room to even place a small version of this loop, and not having good places to attach the top of the loop. I rent, so I can’t really do permanent attachments to the house or carport. So, for my first attempt, I kludged together supports for the end of the wires with 10-foot PVC pipes mounted vertically in slightly larger PVC pipes and a 4-foot wooden dowel inside each long pipe for stability. This was all mounted to an “X” shape cross of two 3-foot or 4-foot pieces of lumber that originally had been used as the cores for my early versions of tuned loops. This was wedged between the house and carport with the loop actually enclosing portions of the carport roof.

There was some sag in the top of the loop, and I kept the bottom of the loop off the ground, so this amounted to something like a 7′ x 25′ loop. When accounting for the multiple turns in a tuned loop, that’s not so much more collecting area. The signals were rather weak, which was probably due to not quite having a proper transformer. The loop also wasn’t quite pointed directly towards any good candidate stations for good back-end nulling. Also, by the time I got this set up, I didn’t have much time left to experiment. But, it was a promising test.

After getting some well-needed advice on a few issues, I tried again. This time, I wound a proper transformer around a toroidal ferrite core, I had a potentiometer with a knob instead of one of those little variable resistors that needs a tiny screwdriver, and I purchased two 20′ telescoping fishing poles to support the ends of the loop. I also left the bottom of the loop on the ground. With wire sag and the over-bendy tips of the fishing poles, this worked out to be something like a 16′ x 30′ loop. That’s actually a bit too high for the length (or too short for the height), but not too far off.

With more than twice the collecting area and proper signal coupling to the radio feed line, I didn’t seem to lack for signal this time. In fact, because my RTL-SDR overloads fairly easily, I had to set the internal gain all the way down to zero to prevent overload, which is primarily caused by my two 1kW locals less than 2 miles from home. The backside of the loop was more or less pointed towards Los Angeles, so I focused on frequencies with stations around there. The best success was on 640 KFI. This generally comes in strong at night with 50kW at 320 miles. It was pretty neat to be able to turn the potentiometer and go from a good signal from KFI, to no KFI and a decent signal on XEJUA out of Juarez from the frontside of the loop at just 5kW and 397 miles away.

This is actually a form of “signal phasing”. It’s somewhat like noise-canceling headphone where an equal but opposite signal is combined with the original signal to cancel out the sound. In this case, you are taking signals from two different parts of the antenna (generally the vertical parts) and due to the non-zero signal path in the horizontal wires plus the effect of a resistor, you get a phase difference between the two signals. With proper tuning of the resistance, you can at least partially cancel out signals in one direction while enhancing signals in the opposite direction. That’s basically why I could hear KFI well at one potentiometer setting, but XEJUA in basically the opposite direction at another setting.

One serious down side is that any particular corner-terminated loop only has a strong null in at most two very narrow angular ranges, and that assumes that one can vary the resistance on either end. (This isn’t as trivial as it sounds, because at the “front” end you also need a transformer to properly couple the loop signal to the cable that carries it to the radio. Thus, you need to have variable resistors and transformers at both ends and a way to switch between them.) If one has a large yard, one can put up several of these loops to null out many specific directions.

However, the term “phasing” is generally reserved for situations where you have two separate unconnected antennas. You can bring those antennas signals together and again use a potentiometer to vary the phasing. If the antennas are identical, that’s all you need. However, if the antennas are identical, they need to be a non-trivial distance apart to act separately enough for good phasing, generally 1/10 to 1/2 wavelength on the lowest desired frequency, with 1/3 being optimal. On the AM broadcast band, 1/10 wavelength at 540 kHz is 55.5 meters (182 feet). There’s a company that sells a ready-to-deploy kit for a phased array of 8 vertical antennas, and it ships with 2000 feet of coax for you to construct the cables leading from each antenna to the phasing system! The optimal array configuration for the 160-meter ham band which has slightly shorter wavelength than the upper end of the AM BCB has the 8 antennas set up in 8 octants of a circle of radius that is 1/3 wavelength, or 175 feet! Anyway, you can buy one of these systems for a little over $3000 and they are ideally suited if you have an open space the length of a football field in both length and width. If one has space in at least a couple directions, it’s common for people to phase two longwire (i.e., a significant fraction of a wavelength) antennas to get a narrow antenna beam in one direction.

However, for those of us who don’t have literally an acre of open space, or even a few hundred feet of linear space to place two identical vertical antennas or two longwires, we have to be more clever. One can get around the 1/10th or 1/3rd wavelength rule by using two different types of antennas. It is typical to combine a loop and a longwire, or a loop and a shorter random-length wire. Even practically right next to each other, they do not affect each other very much and can be effectively phased. When phased together, the combined antennas give a cardioid pattern with a strong null in the direction of the station you are phasing out, and a broad peak in the opposite direction.

Vertical antennas can be problematic if there is a lot of electrical interference nearby, but if not, phasing a tuned loop and a vertical is a good option. I’ve been using tuned loops up to 6 feet diagonal and with the multiple turns of wire adding up to over 100 feet, they give a strong signal at the desired frequency, while the rest of the frequency range is diminished so that I don’t overload my 8-bit RTL-SDR dongle. Luckily, my environment is clean enough to use a decent-sized vertical. After experimentation, I decided on a vertical antenna support made out of two 3-foot wood cross beams, with a 5-foot piece of PVC pipe bolted to this frame. Then, unless it’s windy, I put a 3-foot dowel in the PVC pipe. Finally, a 20-foot telescopic fishing rod goes in the PVC pipe on top of the dowel. It has a ring at the top for fishing line, so I tie the 18-gauge wire to that ring, and voila, a portable 23-foot vertical antenna!

By “portable” I mostly mean that I can easily put it up and take it down every time I am DXing, but this will all easily fit in my compact sport-utility wagon (i.e., “crossover”). Whether I’ll ever take this setup on a “DXpedition” remains to be seen, but at least it’s an option. I would have to drop down to a 4-foot diagonal loop antenna to fit in the car.

For convenience, I set up the vertical about 15 feet away from the loop antenna and run a 25-foot quad shielded coax to the electronics. The big problem I found with “small” vertical antennas is that they really need to be amplified to keep up with the signal coming from a tuned loop. That’s an issue because for ideal phasing, the two signals need to be the same strength. However, I can’t just put a broadband amplifier on the vertical because my two local stations will cause the receiver to overload. The good news is that with an amplified antenna tuner, one can take a broadband antenna and basically do the same kind of “peaking” at a particular frequency as you get with a tuned loop. I’m using an MFJ-1020C active antenna unit, which is really just an amplified antenna tuner with the option of using a small built-in telescopic whip antenna. One removes the whip antenna and plugs a “real” antenna in the back and you are good to go. The tuning part blocks out other frequencies to diminish overloading. The gain is adjustable from -10dB to +10dB, and that upper end is perfect for my purposes. It is important to carefully experiment with gain settings because ideally you want to run as “hot” as possible, without any symptoms of overloading. The RTL-SDR dongles themselves come with an adjustable on-chip amplifier, but with a good pre-receiver set up the dongle gain should be set as low as possible, even though it might go up to >40dB or more.

There are a few commercial phasing units available, but luckily the one that seems to be the most highly recommended for AM BCB is one of the least expensive. Radio Plus Electronics is a small company that offers a limited selection of devices primarily for DXing the AM BCB. Their Quantum Phaser is specifically designed for AM BCB and is particularly well designed. You use two potentiometers to equalize the signal strength off of the two antennas, then adjust the phase difference with another knob. This knob covers about 220 degrees or phase or so, thus there is a toggle switch to immediately flip the phase 180 degrees. This particularly useful because you can adjust the phase to maximize the station you want to get rid of, and then flip this switch to what should be the nulling position. There’s a 3-position switch that lets you toggle between antenna A, the phased signal b/w antenna A and B, and antenna B. This makes it easier to get A and B at the same signal level in the first place, and also lets you see if you have good phasing by checking that the strong station is audible in A and B, but a weaker station is audible at the phased position.

For an example, here is a recording on 1700 kHz. The first 11 seconds is Tejano station KKLF out of the Dallas-Fort Worth area, 917 miles away almost due east and broadcasting with 1kW of power. At 0:11, I flip from the phased position to one of the individual antennas (most likely the loop). You’re then hearing 1700 XEPE out of northern Baja California and they mention sister station “Mighty 1090” (XEPE is an ESPN affiliate and 1090 XEPRS is CBS Sports). XEPE broadcasts at 10kW and is 280 miles southwest of me. They were just finishing a University of San Diego basketball broadcast. This strong XEPE signal is what was on both the loop and the vertical. The KKLF signal (with just a hint of XEPE) is the result of pointing the loop at XEPE and using the phaser to get a ~35dB reduction in the signal from XEPE, leaving a decent signal from KKLF. You can hear me flip back to KKLF at the last second.

It’s not all bread and circuses. Getting the best possible null requires very precise adjustment of the antenna signals and the degree of phasing. You adjust the antenna gains, then get the best possible phasing, then adjust one of the gains, then adjust the phasing again, etc. Plus, you need both the loop capacitor and the pre-selector tuned to as precise of frequency as possible. Also, the loop needs to be as close as possible to pointing directly at the station you are trying to null out. The stronger the station, the more important it is to have all the stars aligned to get the best possible nulls. A phaser can also be quite useful with weaker stations

**For some perspective on the “soil” conditions in my area, the city was in the process of rebuilding a street a block away, including installing large drainage pipes under my street to carry away the occasional 1-inch-per-15-minute monsoonal deluges we get in the summer. Right in front of my house, they had to use explosives to clear an outcrop of hard granite a few feet below the surface. They broke metal drill bits and failed at breaking it up with mortars. One day, I left a pocket voice recorder by the side door, not in the direct line of sight to the explosion site. Here’s the final warning signal, followed by 40 seconds of silence, a verbal warning at 1:05, the explosion, and finally the “all clear”:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s