The Dark Ages of Quad Design

he cubical quad - our prototypical loop antenna - has been on the scene since its discovery by Clarence Moore in 1942. Until just the last couple of years virtually nothing has been done to analyze its performance and improve upon it. Note however, when I speak of "quad" antennas, I am using the term generically for the entire class of loop-based antennas. The square loop was the simplest and the first to be used.

Yet, quads are popular for two reasons. First, they are cheap and construction materials are readily available throughout the world. Spreaders can be of bamboo or fiberglass. Virtually any type of wire has been used, including the commonly-recommended #14 AWG for HF antennas.

Second, there are claims of superior performance when compared to Yagis. It is alleged that a quad-based parasitic array can significantly outperform a much larger Yagi having more elements. Or, that a quad of "X" elements will have the gain of a Yagi of X+1 elements. Other authors have claimed that the quad has an advantage due to its lower takeoff angle (TOA) when compared to a Yagi at the same height.

The performance claims, not the limitations of money, have lead competitive amateurs over the years that I have been hamming (since 1957) to put up elaborate multi-element - up to 5 or 6 - monster quads in order to acquire a gain advantage over their brothers (rivals, contesters, DX'ers). It wasn't until Wayne Overbeck, N6NB, in the early 1970's, took a portable crankup tower to test antenna gain at various stations with large arrays that the "Quad vs. Yagi" controversy was resolved - in favor of the Yagis. Not only were the quads' gains inferior to Yagis with the same number of elements but they were not as rugged. The takeoff angle myth was also debunked. The quad may have less high-angle radiation than a Yagi but it has no greater radiation at low angles.

So now that we have known, for almost 30 years, that the quad has no advantage in gain or TOA over a Yagi, have we seen the design thoroughly analyzed and improvements published in the amateur or antenna literature? No, not at all. At least not until L.B Cebik, W4RNL began his studies in the last couple of years. This almost 30 year period of stagnation is what I refer to in the title as the Dark Ages.

During these years Yagi design has progressed from systematic analysis in the original NBS designs of the late 60s and Jim Lawson's, W2PV, detailed studies in the late 70s to readily available computer optimizing programs at the present time. In fact, the OWA (Optimum Wide-band Array) design was computer generated.

Meanwhile, what has happened with quad design over the last 30 years? Virtually nothing. There are antenna books and handbooks written for hams in many languages which continue to propagate the same dimensions and analyses over and over again. There is a chapter in the last Antenna Handbook in which the author says that element spacing does not mean very much. Run a Web search on "quad" and "antenna" and you will come up with many sites promising to give you exact dimensions for any frequency and which will feed you JAVA applets purporting to calculate these dimensions-all the while basing the "pseudo-science" on the old erroneous formulas.

The major variable in quad construction is the wire diameter and, as L.B Cebik stated in his recent article in antenneX, absolutely none of the formulas take this into account. Meanwhile, the differences due to wire diameter are truly gross-a quad based on #10 AWG (2.5mm) wire has a loop perimeter of 1.046 wavelengths on 80 meters and 1.062 on 10 meters. If you think these decimals don't mean very much, the difference in perimeter between these two figures on 80 meters is 1.34 meters or 4.4 feet. Now extrapolate the same wire size to 2 meters. The perimeter is 1.084 wl.

When it was suggested I write an editorial for this month, I asked LB where to look for more articles and papers on quad design. Who should I contact? His answer was that he, David Jefferies, G6GPR, and I were doing the pioneer work in amateur radio. In addition Bob Haviland, W4MB continues his search for basic quad properties, extending the material in his 1993 book, in a delayed series of articles in QEX (begun in Communications Quarterly). I trust LB's observation about this since he has been around far longer than I and is more knowledgeable about the field. I had not thought of myself as a pioneer, just as a dilettante who has played around with loop-derived antennas for about three years.

As an aside and as a point I will come back to later in this editorial, I got started in this field by accident. I was looking for an antenna for 80 meters that I could erect in my yard and which had more gain and better low-angle radiation than a dipole. I came across LB's website and found the rectangle. This seemed like the ideal antenna. Its radiators were vertical and not too tall. In fact, in this vertically polarized form it works better when fairly low to the ground, and its horizontals were about the length of a dipole.

At about the same time two other things happened that planted some seeds in my brain. First, I got started in antenna modeling. Second, I read an article about the Hentenna which is a tall horizontally polarized rectangle intersected at some point by a third radiator. In effect it was two loops, of different sizes, attached to each other. Just for the fun of it I decided to model the Hentenna as one of LB's rectangles - turned 90 degrees so that the radiators were vertically polarized and the long wires horizontally oriented.

I had a lot of time to play, and play I did, because I was incapacitated with back problems. Fortunately, these injuries were resolved by two surgeries but I had been bitten by the bug. I didn't have to do very much to come up with superior designs to those in the literature. Some time later, I had some interesting results but couldn't explain them. Again, I was lucky. I found David Jefferies through his website and got him interested. He, being a Ph.D, and a Professor of Electronic Engineering in the UK could explain the phenomena that appeared anomalous to me.

What do we KNOW about quad design?
Wire thickness makes a BIG difference:

The thicker the wire the greater the loop perimeter. The greater the perimeter the higher the loop impedance, the lower the losses, the higher the gain and the lower the Q factor. People normally construct Yagis out of thick aluminum tubing. They normally construct quads out of thin copper wire. Then they expect the quad to outperform the Yagi.

Quad elements couple better:
Loops of whatever shape and number (where number refers to the number of loops enclosed within a single element) couple strongly to other loops. Rod elements, in Yagis, do not. It is very easy to get equality in the loop currents in a parasitic 2-element quad and thus attain a tremendous front-to-back (f/b) ratio. As David Jefferies has explained, the coupling is via the corner currents and the non-radiating wires which connect the radiators. With respect to 2 element antennas the quad's lowest f/b at the band edges (for any band besides 10 meters) is higher than the highest f/b at the design center frequency of a Yagi.

The lesson we learned from this is that the ideal element spacing in quads has to be far wider than published in the old, stale, literature. Many benefits accrue from proper coupling. Besides the higher f/b ratio, we end up with a far greater BW and an equal division of currents. The current division is important since the losses of such an antenna, especially on the low HF bands, are a function of the square of the element currents. In the case of parasitic rectangles, we can increase the gain significantly on 80 meters simply by using the proper element spacing and reducing the loop losses. We can also almost double the SWR bandwidth as compared to that of a single loop.

The spurious "Q":
As LB Cebik has pointed out, one of the selling points of quads is that they are lower in Q than Yagis. Their SWR BW is far wider. This is misleading however since the f/b BW is far more restricted than the SWR BW. This shows up on 10 meters where a thin-wire quad's useable f/b (15+ dB) BW covers less than one half of the 1.7 MHz (6%) wide band.

New ideas:
Improvements in loop antenna design are coming from two directions. LB is improving the design of simple quads by using two parallel wires in place of the single wires. If spaced far enough apart they act as if they are a much thicker wire. This results in higher gain, a more stable gain across the entire BW and a much wider f/b BW than ordinary quads.

On the other hand, I am playing with the "multi-loops". These are individual elements composed of more than one loop. My generic term for the Hentenna, one which is composed of two different-sized loops, is (at LB's suggestion) the ADR or asymmetrical double rectangle. The old magnetic slot (composed of two equal-sized loops) has a generic name of symmetrical double rectangle or SDR. Added to all of this are the triple, quadruple, .... n-loops. These antennas offer higher gain and a much wider useable f/b BW.

Both design pathways involve a little more complexity in construction as a tradeoff for much improved performance. All the antennas are easily constructed with the same materials which have made the quad so popular where aluminum tubing for Yagis is unavailable or very expensive. Even when money is no object but "killer" performance is, the newer loop designs offer the performance of much larger Yagis - but on a shorter boom.

My objectives with this editorial:
This field is wide open for anyone with a little patience and some ideas. It is obvious that, with respect to loop antenna design, we are just coming out of the Dark Ages. There is a tremendous room for improvement. It is also obvious that you don't have to be an antenna engineer. The Web is truly wonderful; there are knowledgeable people willing to help.

As I pointed out, I was the least knowledgeable person with respect to these antennas and antenna modeling. I am a true "amateur" and my only knowledge base was from what I had gleaned from reading magazine articles and antenna books. Yet I learned something useful over the last 3 years. (I was even forced to relearn college calculus). These antennas are ripe for the picking by anyone out there with a little interest and a modeling program.

I am hoping to spark some interest among the experimenters out there. I hope we can get loop design to the same stage as Yagi design, but in a much shorter time span. If you have an idea try it out and publish the results here. -30-

**Author's Brief Biography**
Dan Handelsman, N2DT Dan Handelsman, N2DT was first licensed as WA2BCG in 1957at age 13. He became interested in antennas at that time when he had to figure out a way to operate from the 6th floor of his apartment house. This resulted in a mobile whip being stuck out from a window without a counterpoise. At that point he became an "expert" in TVI. He was licensed as N2DT in 1977 and is a DX'er and contester. He is now playing with experimental antennas and low power. Professionally, he is a Pediatric Endocrinologist and holds M.D. and J.D. degrees and is Clinical Professor of Pediatrics at the New York Medical College. As far as his antenna work he is an "amateur" in the truest sense of the word (Dan's words!).