K9AY Loop


OK, I put up my K9AY Loop and it has very poor front-to-back. Everything seems to be connected properly. What can I do?

We have worked with many users to find problems — here are the most common:

1. Loop wires connected to the wrong terminals — be sure both ends of the same loop are connected to the matching transformer and termination!
2. Control wire problem — a broken wire (check inside the insulation), a mis-wired connector, and possibly too low voltage to activate relays. Be sure to do all control wiring work with the power off to avoid damage due to accidental shorts.
3. Grounding problem — Poor connection to the ground rod, coax shield loose in its connector, too short ground rod in dry soil.
4. Coupling to nearby metallic objects — Towers, other antennas, metal fencing, power lines, or the metal siding on nearby building (don't forget the metal lath under stucco!).


Do I need radials under the loop to get a better ground?

In most locations with clay or loam topsoil, the ground rod is the only ground system that is required. Many sandy soil locations are also OK is they are not in desert areas.

Based on reports from users, radials are only needed in extreme or unusual locations. Desert locations in particular seem to benefit from an improved counterpoise. Elevated installations may also benefit from radials (see FAQ below on this subject).


Can I install the loop on my roof?

Only as a last resort. The K9AY Loop requires a ground, just like a vertical antenna. This allows it to have convenient direction switching. Users report mixed results with installations above ground. The antenna workes best on flat ground, but can handle a few feet of height with little or no change. For example, users have raised the antenna so that its base is even with the top of a metal fence. Raising the antenna 2 or 3 feet can make mowing under it a lot easier, too.

Rooftop mounting has had mixed results at best, according to user feedback. There have been successful installations on an aluminum sunroom roof, and at the edge of a roof where the wire to the ground rod is nearly vertical. We have have one report each of the antenna working well on a flat metal roof, and on a 40-foot tower, installed above a tribander. However, we have also had several users tell us that nothing they tried would make a rooftop antenna work.

We recommend other antenna options for rooftops — a pennant or flag will usually work, since they do not require grounding, although sometimes the reflections from a metal roof or metallic objects in the building below will disturb those antennas, too.


How far should the K9AY loop be from my transmit antenna?

As far as possible — My own loops have worked just fine when located just beyond the radial system of an Inverted-L (100-125 feet). Users have reported using the loop as close as 50 feet (high power) and 20 feet (500 watts), only with the addition of protective circuitry that disconnects the antenna from the radio when transmitting.

The other factor is coupling to the TX antenna. This is a bigger problem with a shunt-fed tower than with an inverted-L. It appears that most inv-L antennas are sufficiently de-tuned when the T/R relay is in the receive mode. Overhead dipoles and inverted-vees can be a big problem, but if luck is with you, they are also detuned during receive by the open-circuit coax between the antenna and T/R relay.


Can I make the loop larger, smaller or a different shape?

The short answer is yes — many hams have limited space or want to use existing supports such as trees. Here are the general rules about loop size and shape:

1) The current in the loop (and thus, signal voltage across the feedpoint terminals) is proportional to the area enclosed by the loop. — A smaller loop will capture less signal, which will require more preamplifier gain to give the receiver a proper signal level. The loop is too small when the quiet band signal level falls below the noise floor of the receiving systen (preamp/feedline/receiver).
A larger loop will capture more signal, but at some point its mode of operation changes. A loop is too large when it starts to act like a "delta loop" or "quad loop" instead of a "small loop." A loop is too large when the overall wire length is about 0.3 wavelength; approx. 0.1 wavelength across the diameter or longest diagonal dimension. The published dimensions (25 ft. height, +/- 15 ft. across) are right at the upper limit for operation in the 80 meter band. Also, as the size approaches the upper limit, the optimum terminating impedance will change, and the maximum front-to-back will start to degrade.

2) The shape of the loop determines the vertical angle of the null. — The semi-delta loop shape was chosen for two reasons -- it is practical for a single tall support, and it places the null at an elevation of 35-45 degrees above the horizon, which is ideal for rejecting the strong signals coming from one-hop skip distance.
A square or rectangular shape with vertical ends will place the null at a lower angle, which may be better for local noise rejection. A diamond or a tall rectangle that is higher than its width will place the null at a higher angle, which may be useful for rejection of very short skip (NVIS) signals.


How do I use the variable termination?

In the majority of installations, you will only use one or two settings of the termination. For a wide range of soil types, you could install a 390 ohm resistor and forget it! But, for perhaps 10 or 15% of users, an adjustable termination is needed.

In extreme or unusual locations (desert, rocky terrain, seashore...), the optimum terminating resistance is usually different than it is for typical soil. As noted in the instructions, do plenty of listening to determine what settings work best. Check the front-to-back at various resistances while listening to WWV on 2.5 MHz, CHU on 3.330 MHz, AM radio stations above 1400 kHz, or W1AW transmissions on 160 and 80 meters.

Remember, the termination does NOT tune the antenna — it only optmimizes the depth of the rearward null. And that optimum setting does not change much. As designed, setting the termination for optimum front-to-back in the 160 meter band will maintain that optimum F/B at all lower frequencies, while 80 meters may (or may not) require one step higher resistance.


What about using a Vactrol for continuous resistance adjustment?

First, the antenna does not require a precise resistance for termination. The 8-steps provided (from 340 to 680 ohms) are sufficient to get the deepest rearward null.

Vactrols may be useful in certain circumstances, but have two significant limitations — They have very low power dissipation (less than 1/4 watt) that makes them susceptible to damage from transmit power and static charges. Also, they have a lot of drift with time and temperature. Reports of "tuning" the K9AY Loop with a Vactrol are almost certainly the result of re-adjusting the Vactrol to the optimum resistance value after it drifted off.

A widely-adjustable termination like the Vactrol (or a poteniometer, as some users prefer) will be most useful for SWLs who set up the loop in different locations for special listening DXpeditions. It will help accommodate the wide range of possible ground conditions and operating frequencies.


Can I make one rotatable loop?

Of course! On the plus side, it will allow you to place the null in any direction. However, you will lose the instant switching ability of the crossed-loop design, which is not only useful for front-to-back testing, but lets you quickly react to signals arriving from different directions.


My loop has an SWR of 2:1 is this bad?

No — for a receiving antenna, we usually do not care much about SWR, unless it is so high that there are excessive losses in the system. In normal operation, the SWR at the K9AY Loop feedpoint will be in the 1.5:1 range, but that can change with termination value, ground conductivity, and frequency. If the SWR is higher than 3:1, it may indicate a problem — an inspection and ohmmeter check of the system may be in order.

If SWR is critical for your use (e.g., for precise control in an array of multiple loops), advanced users can optimize it for one band. The impedance of the antenna is established by the terminating resistance, through the inductive transmission line effects of the loop wire. A single capacitor in series with the loop wire where it connects to the feedpoint (not the termination) can compensate for this inductance on one band. A more complex network is needed for wideband or multiband compensation.


Since you mentioned arrays, how does that work?

The K9AY Loop is ideal for larger arrays, since it has already has a directive pattern. For example, two K9AY Loops will have a far better directive pattern than is possible with two omnidirectional verticals. There are hams who have built 4-squares of K9AY Loops and marveled at the extreme front-to-back performance.

We have done experiments with arrays of K9AY Loops and expect to offer materials to allow others to build them. Note that these will not be as compact as a single loop, but these arrays are all smaller than Beverage antennas with equivalent performance and will be good options for hams who have some space available, but not enough for Beverages.


What about separate antenna and radio grounds for common-mode rejection?

The switching circuitry is simplified by using a single ground for the antenna, control voltages and receiver output (coax). This was done after determining that there is low liklihood of common-mode problems. First, most K9AY loops are installed with relatively short coax and control lines. This is a situation that has low susceptibility for common-mode problems unless the feedline is elevated above ground (more about this later). Also, the small size of the loop means that it will not pick up a lot of signal in the common-mode, or "short vertical" mode.

But common-mode CAN happen in some cases, which will degrade the directivity of the antenna. A very long coax and control line can introduce significant common-mode signal pickup. Similarly, any elevated feedlines are efficient, and unwanted, antennas.

The first defense against common-mode problems is to bury the coax and control lines. Laying them on top of the ground is the second best choice. The lossy dielectric earth will attenuate the common-mode energy. If you believe common-mode is still a problem, we recommend that you wind a few turns of both the coax and control wires through a large, high-mu ferrite toroid. This should first be done near the antenna, with a second such choke placed near the shack if the problem persists. Remember, whenever you use such a choke, an additional ground will be needed. In some cases, a ground at the entrance to the shack will suffice, making a total of three grounding points. If the second choke is used, it should go between this ground and the radio.