Wide-Band 40-Meter Yagis
Part 2: Alternative 4-Element Designs

L. B. Cebik, W4RNL (SK)

In Part 1 of our expanded notes on wide-band 40-meter Yagis, we looked at some standard and non-standard Yagi design ideas. The 2- and 3-element Orr-derived Yagis employed wide element spacing to accomplish 2 goals: to raise the general feedpoint impedance to the 50-Ohm level and to increase the operating passband to cover all of 40 meters with under 2:1 SWR. However, only the relatively long-boom 4-element Yagi that used a master-slaved dual driver system achieved the most desirable end of having well under 1.5:1 SWR across the band while raising the gain level to about 8 dBi in free space. We also examined the recent N4HBX reflector-less design that employed a fed driver and bracketing slaved drivers to obtain about 7 dBi gain with under 2:1 SWR across the band.

In the interest of making fair comparisons among the designs, we employed the same element-diameter schedule in all elements of all beam designs. Fig. 1 replicates the taper schedule. As we proceed through the new notes, the structure tables will list only the element spacing (using the reflector as the baseline), along with the total element half-length and the tip length. You only need to add the 3/8"-diameter tip length to the schedule in Fig. 1 in order to replicate any model within the series.

As we closed the first part of these notes, we noted that we had not exhausted all of the design options that we might apply to Yagis in order to achieve full band coverage. In the following notes, we shall take up the design challenge in 2 directions. One path uses direct-coupled dual drivers as a substitute for the open-sleeve coupling system used in two of the Part-1 designs. The other roadway leads to phased pairs of elements used as drivers. In both directions, we shall restrict ourselves to 4-element designs that produce between 7 and 7.4 dBi gain with a good front-to-back ratio across the band. We shall aim at both shorter and longer versions of each design both to display the performance differences and to once again stay below 1.5:1 50-Ohm SWR from 7.0 to 7.3 MHz--if possible. A shorter Yagi version will be one that is between 31 and 36 feet long, while a longer version will be between 36 and 42 feet. All arrays will be shorter than the 45' boom required by the standard 3-element wide-band Yagi that also yields forward gain in the same low-7-dBi free-space range.

Before we move into precise design details, however, we should look more closely at all of the varied Yagi driving schemes for widening the operating bandwidth of a beam. Each type of driving system achieves its goal by different means.

How Drivers Drive a Yagi

A standard Yagi has a single driving or fed element, along with a one or more parasitic elements. By convention, any element behind the driver--relative to the direction of the antenna's main lobe--is a reflector, and any parasitic element ahead of the driven element is a director. In an open-sleeve coupled driving system, we directly feed only one element. Hence, various sources may call the slaved driving element either part of the driver or a director. Either title is technically correct, since a slaved driver obtains its energy parasitically. It becomes a driver by dominating the current magnitude from which other elements derive energy parasitically. In the notes that immediately follow, I shall identify the driving elements, whether directly connected to the source or not.

In most standard designs, the reflector and director elements are sufficiently distant from the driven element to qualify strictly as parasitic but not driving elements. Again, the distinction is more conventional than strictly technical, since every parasitic element contributes to forming the antenna's radiation pattern. Let's consider the 3-element standard wide-band Yagi from Part 1. If we provide the driver with a relative current magnitude of 1.0, and then, if we move the antenna across the band, we find that the current magnitude levels on the parasitic elements changes along the way. The following table shows the relative current magnitudes on all of the element centers at 7.0, 7.15, and 7.3 MHz.

Relative Current Magnitude on the Elements of a 3-Element Wide-Band 40-Meter Yagi
Element 7.0 MHz 7.15 MHz 7.3 MHz
Reflector 0.385 0.289 0.219
Driver 1.0 1.0 1.0
Director 0.525 0.625 0.734

The table does not show the phase angle of the currents. The driver, by model set-up, is always at 0-degrees. The reflector normally shows a positive phase angle that falls anywhere between 100 and 150 degrees relative to the driver. The director phase angle is normally negative relative to the driver and may be anywhere between 80 and 150 degrees. The exact phase angle for each parasitic element derives largely from the position and size of each element relative to the driver. If you see only a current table for a set of elements without designations such as "reflector" and "director," you can still tell the direction of the main beam by looking at the phase-angle entries.

Fig. 2 replicates the data in the table using a more graphical presentation. The pink vertical lines represent the relative current magnitude on each element at each sampled frequency. The figures may make the progression of current values more readily apparent.

Very noticeable in the table for the standard 3-element Yagi is the fact that as the frequency increases, the relative reflector current magnitude decreases. In Part 1, I noted in connection with certain designs that the reflector size has a greater effect on lower frequency performance than on higher frequency performance. Equally, the director size has a larger effect on the higher-frequency performance. As the table shows, the director current magnitude increases as we raise the beam's operating frequency.

The pattern of current magnitudes for the 4-element open-sleeve coupled design in Part 1 shows a similar progression of current values for the reflector and the most forward element, which I shall simply call the director. The exact values differ from those of the 3-element Yagi because the element lengths and position differ between the two designs. The center 2 elements form the master and slaved driving elements, and we should pay close attention to them. Since we feed only one element, its current magnitude is always 1.0. In the following table, note the relative current magnitude on the slaved driving element as we increase frequency. Also follow the current magnitudes in the graphical representation in Fig 3, which appears just below the table.

Relative Current Magnitude on the Elements of a 4-Element Wide-Band 40-Meter Yagi
Element 7.0 MHz 7.15 MHz 7.3 MHz
Reflector 0.569 0.413 0.276
Master (Fed) Driver 1.0 1.0 1.0
Slaved Driver 0.683 0.909 1.120
Director 0.435 0.544 0.646

As we increase the frequency, the current magnitude on the slaved driver reaches and then exceeds the current on the fed driver element, the one to which we have connected the feedline. (Remember that the current levels are relative to an assumed value of 1.0 at the feedpoint.) If we were to remove the slaved driver, then, at 7.3 MHz, the current level on the director would fall considerably--or its phase angle would change radically--with the net result that the beam's performance would decline in one, and usually several, categories. In designs like this one, we tend to cut the directly fed driver long to obtain good low-end performance and then to cut the slaved driver high to extend the performance at the high end of the band. If we carefully select the reflector spacing, we obtain a very wide SWR curves with a low value across the band. Essentially, the slaved driver's control of the beam's pattern at the higher end of the band--where the relative current equals or exceeds the driver current magnitude--extends back to the feedpoint so that the impedance at the junction with the transmission line remains stable. You may recall that the 4-element design in Part 1 showed a 50-Ohm SWR that did not grow to 1.25:1 anywhere within the 40-meter band.

Many classic Yagi designs have used the open-sleeve dual driver system without realizing that fact. Early literature on the famous DL6WU VHF/UHF Yagi series used a reflector-to-driver space of 0.2 wavelength and a driver-to-director-1 spacing of about 0.05 wavelength. The result was a wide-band Yagi that covered all of the 70-cm band with an acceptably low SWR. Current analyses of the driver and first director elements show that the two form a master-slaved driver pair. Other well-known designs have followed this lead, using other spacing combinations. In the HF range, the OWA (optimized wideband antenna) Yagis used the same type of driving system, with different relative element spacings. However, the OWA series of beams also used special treatments of the second and third directors that stabilized a number of other performance properties. Hence, a Yagi may be a wide-band Yagi by using open-sleeve coupling between a fed driver and the next forward element without being an OWA Yagi. Since we shall not have enough directors to provide the special OWA treatment, all of our examples will simply be wide-band Yagis.

The N3HBX reflector-less Yagi uses a 3-element driving cell, with slaved driving elements behind and ahead of the fed driver. Since the driving cell is an example of open-sleeve coupling among the elements, we should expect a pattern of changing current on each element, with the rear driver showing reduced current and the forward driver showing elevated current as we increased frequency. The following table and Fig. 4 confirm our suspicions.

Relative Current Magnitude on the Elements of a 4-Element N3HBX Wide-Band 40-Meter Yagi
Element 7.0 MHz 7.05 MHz 7.15 MHz 7.25 MHz 7.3 MHz
Rear Slaved Driver 0.985 1.014 0.686 0.422 0.221
Master (Fed) Driver 1.0 1.0 1.0 1.0 1.0
ForwardSlaved Driver 0.520 0.930 1.092 1.541 1.282
Director 0.286 0.458 0.481 0.625 0.433

The relative driver current magnitude levels do not disappoint our expectations. I have added 2 extra columns to the table because we noted that the impedance and SWR curves for the Yagi design showed peak performance values just inside each band-edge. The current values reflect the performance data. At 7.05 MHz, the rear slaved driver has close to its highest value. Likewise, the forward driver shows nearly peak magnitude at about 7.25 MHz.

Many antenna designers use open-sleeve coupling for multi-band antennas, tuning each slaved driver to a band that is usually higher in frequency than the band to which they tune the fed driver. The technique has seen successfully use in both vertical monopole systems and in multi-band Yagis. One of the earliest conscious in-band uses of open-sleeve coupling involved the design of a dipole to cover all of the 80-75-meter band, or at least selected portions of the band.

An alternative to open-sleeve coupling for multi-band antennas is direct coupling, sometimes called closed-sleeve coupling. Optimbeam Yagis use such a system, and the Bencher Skyhawk changed its feed system from open- to closed-sleeve coupling a few years back. One advantage of direct coupling over open-sleeve coupling is a broader bandwidth on frequencies higher than the lowest covered by the antenna system. The principle of direct coupling is simple-sounding, but electrically somewhat complex. Let's take a pair of driver elements. Let one be connected to the main feedline. Between that element and the other, we shall make a direct connection with a fairly low-impedance transmission line. The line will be short and not reversed. The close proximity of the two drivers produces parasitic or high mutual coupling similar to that found in open sleeve coupling system. However, we have also fed power directly to both elements. One result is that directly coupled drivers usually use element spacing that is greater than for driver pairs that use only parasitic coupling. A second result is the need to carefully balance the connecting transmission line impedance with the element size and spacing.

Most direct-coupled driver system emerge from computer experimentation, which is far more rapid than physically varying the components with a prototype antenna. When successful, the multi-band driver set--amid the other elements forming a Yagi--shows a good match to the main feedline on each band. However, where we connect the feedline may vary, with some feedlines connecting to the higher-frequency element and some to the lower-frequency element.

When we translate a directly coupled driver pair from multi-band to in-band service, we may encounter some surprises in the sense of needing a configuration unlike those that we saw in open-sleeve coupled driver systems. The following table and Fig. 5 track element currents for a shorter 4-element 40-meter wide-band Yagi that uses a pair of directly coupled drivers, with a short, direct 50-Ohm transmission line between the driving elements. Note from the graphic that the feedpoint is on the rear driver of the pair. This driver is also the shorter of the two interconnected elements.

Relative Current Magnitude on the Elements of a 4-Element Wide-Band 40-Meter Yagi
with a Directly Coupled Driver Pair
Element 7.0 MHz 7.15 MHz 7.3 MHz
Reflector 0.505 0.395 0.300
Rear Driver 0.194 0.144 0.300
Forward Driver 0.780 0.935 1.357
Director 0.398 0.548 0.783

The rear driver appears to carry a very low current magnitude, but the directly coupled pair yields a large operating bandwidth--as large at least as the bandwidth of the long-boom 3-element Yagi, but on a boom that is 10' shorter. How the directly coupled pair operate in this particular design is quite interesting. The phase angle of the rear driver is positive and almost as large as the positive phase angle of the reflector. The parasitic reflector element shows a decrease in relative current magnitude as we increase the frequency, but the rear driver shows an increase in current over that same span. The net effect is fairly smooth performance over the entire 40-meter band. Since we have a division of current at the main feedpoint, neither element shows a relative current magnitude of 1.0. As well, we also have changes in the phase angle from the feedpoint or source value of 0-degrees. Since even a short low-impedance transmission line will show impedance-transforming properties, the driver magnitudes do not add arithmetically. The directly coupled driver pair used in this design exercise is not the only one that we might employ, but it is satisfactory to our goal of producing a wide-band 40-meter beam.

Our final example involves the use of a driver pair arranged to form a system traditionally called phased feeding. The two driver elements are more widely spaced and use a reversed (or half-twist) transmission line section to achieve proper phasing. Phasing does not directly concern the feedpoint impedance at each element in the pair. Rather, we arrange the element lengths and their spacing, as well as the line impedance between them, to achieve a desired relative current magnitude and phase on each element. Since the phased pair of elements forms a 2-element LPDA arrangement, the resultant beam that we produce when we add a reflector and a director often bears the name "log-cell Yagi." However, for small log cells (fewer than 4 elements), the driver pair usually emerges from experimentation rather than the application of LPDA equations. The systems used in our wide-band 40-meter designs employ a 250-Ohm phase line, and the feedpoint is on the forward element of the pair. The following table and Fig. 6 provide data on the shorter of the two versions of the wide-band log-cell Yagi that we shall later examine in more detail.

Relative Current Magnitude on the Elements of a 4-Element Wide-Band 40-Meter Yagi
with a Pair of Phased Driver Elements
Element 7.0 MHz 7.15 MHz 7.3 MHz
Reflector 0.223 0.155 0.102
Rear Driver 0.654 0.655 0.453
Forward Driver 0.674 0.824 1.030
Director 0.610 0.905 1.163

The reflector shows decreasing relative current, while the director shows increasing relative current as we increase the operating frequency. The rear driver trend is downward, while the forward driver trend is upward. However, perhaps the most notable aspect of the system is the relative unimportance of the reflector and the very high importance of the director in terms of their relative current levels all across the band. We can remove the reflector and sustain the feedpoint impedance at nearly the 50-Ohm level and the overall operating bandwidth. The director plays the most important role in setting the radiation pattern behavior. You may relate this discussion of relative current levels to the Part-1 discussion of reflector-less Yagis using a phased pair of driver elements.

Although our excursion into Yagi element currents has delayed our introduction of additional wide-band Yagi designs, it has shown us a few things, and has potential for showing us others. The diversity of parasitic arrangements by which we can achieve a set of operating goals within an overall beam structure continues to grow as designers bring their ingenuity to bear on their challenges. An understanding of the current patterns that emerge as we track a design through its operating range contribute to our better understanding of the design, even in the absence of explicit design equations. In fact, we might even use the information that we have just surveyed to understand some of the behavioral quirks in the designs to come.

The Design Challenge

In Part 1, we examined a 3-element standard Yagi on a 45' boom. It provided the performance listed in the following table from that discussion.

Frequency       Free-Space       Front-Back        Feedpoint Z               50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 6.92 19.16 51.6 - j 17.6 1.41
7.15 7.03 21.54 48.9 - j 1.3 1.04
7.3 7.37 19.00 41.6 + j 18.8 1.57

The Yagi averages about 7.1-dBi forward gain and close to 20-dB front-to-back ratio. It shows a rising gain curve as we increase frequency. The 50-Ohm SWR curve is very shallow and just misses the 40-meter grail of being under 1.5:1 at the upper end of the band. (We shall not here debate the merits of calculating the SWR at the end of some arbitrary length of lossy coaxial cable.) Boomlength remains the key drawback to the design and is not eliminable. A shorter boom would narrow the operating bandwidth and lower the resistive component of the feedpoint impedance, thus increasing the SWR over a large part of the band.

In contrast, the 4-element N3HBX design requires only 31' of boom. The triple-driver section with a single director produces the performance shown in the following table, repeated from Part 1.

Frequency       Free-Space       Front-Back        Pre-Match Z        25-Ohm        Post-match Z       50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR R +/- jX Ohms SWR
7.0 7.21 20.91 16.3 - j 2.8 1.57 67.7 + j 20.5 1.59
7.15 7.18 25.20 31.7 + j 0.4 1.27 38.7 - j 0.1 1.29
7.3 7.46 20.69 15.7 + j 4.8 1.69 72.6 - j 20.4 1.65

Forward gain averages about 7.25 dBi, with a 22-dB front-to-back ratio. The average feedpoint impedance is in the vicinity of 25 Ohms. So the array requires a matching network. The quarter-wavelength section of 35-Ohm transmission line yields the final columns of impedance and SWR values. Although the antenna does not produce as low a set of SWR values as the 3-element array, it remains well under 2:1, with a savings of 14' of boom but a cost of the weight of one more element.

Obviously, the challenge facing any competing design is to improve upon one or more sets of figures while maintaining the shortest possible boomlength. We shall look at 2 different ways of approaching the challenge.

4-Elements and Directly Coupled Drivers

The first design direction involves the use of a driver pair with directly coupled elements. A length of 50-Ohm transmission line connects the 2 drivers. The rear driver is the shorter one and also serves as the feedpoint for a direct connection to 50-Ohm cable. To this pair of drivers we shall add a reflector and a director. The following table provides dimensions for the shorter version of the 4-element wide-band Yagi with directly coupled drivers.

Dimensions of the 4-Element Wide-Band Yagis with Directly Coupled Drivers

Note: All dimensions are in inches. Multiply by 2.54 for centimeters,
0.0254 for meters.

Antenna Element Spacing Half-Length Tip Length
Short Reflector 0 461 75
Rear Driver 112 397 11
Forward Driver 177 425 39
Director 2 418 386 0
Long Reflector 0 461 75
Rear Driver 192 398 12
Forward Driver 257 425 39
Director 2 498 386 0

The short version of the array has a boomlength just under 35'. Within those limits, it produces the performance figures that appear in the following table.

Frequency       Free-Space       Front-Back        Feedpoint Z               50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 7.05 22.35 28.4 - j 7.1 1.81
7.15 7.07 24.54 36.1 + j 2.6 1.39
7.3 7.31 22.06 45.2 + j 22.0 1.60

The curves in Fig. 7 translate the sample values into a full-band picture of performance. Below the graphs are free-space E-plane patterns at the sampled frequencies.

The gain and front-to-back figures are very comparable to those for the N3HBX beam without some of the dimensional finickiness that goes along with the triple-driver that uses a master and 2 slaved driving elements. The directly coupled system provide well under 2:1 50-Ohm SWR on a boom only 4' long than the N3HBX boom, but it does not result in the desired 1.5:1 50-Ohm SWR maximum.

Since the resistive component of the impedance values is uniformly low, we can increase it by increasing the distance between the reflector and the rear driver. As we increase this distance, the feedpoint resistance does not rise as fast as it might with a single driver. As noted earlier, the rear driver exerts reflector-like influence on the radiation pattern due to the positive phase angle of the element-center current, an angle that almost equals the phase angle of the reflector current. Therefore, we must increase the distance by nearly 7' to achieve the desired SWR curve. So the final boomlength of the long version of the 4-eleent directly coupled driver Yagi is about 41.5'--still short of the boomlength for the simple 3-element wide-band Yagi. The following table samples the performance.

Frequency       Free-Space       Front-Back        Feedpoint Z               50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 7.13 21.01 36.9 - j 10.9 1.48
7.15 7.19 24.23 42.1 - j 1.5 1.19
7.3 7.42 21.59 49.6 + j 18.8 1.45
Fig. 8 converts the sampled data into full-band curves and provides the e-plane patterns.

The only changes in the long version, relative to the shorter one, involve moving the reflector and changing the length of the rear driver by 2" (1" per half-element length). Since the director position and size do not change, the radiation pattern performance does not change, although the gain does increase by a paltry 0.1 dB. However, the array does not exceed a 50-Ohm SWR value of 1.5:1 across the whole band. Although shorter than the 3-element Yagi, the array is considerably longer than the N3HBX beam (by 10.5'). However, the long directly coupled Yagi is about 10' shorter than the original 4-element Yagi with its added dB of gain and its impressive SWR curve.

The use of a directly coupled driver pair puts the potential builder in a quandary. Fig. 9 compares the outlines of the 2 beams and reveals the key difference: the driver-to-reflector spacing needed to achieve the better SWR performance. At the short length, and if a 2:1 SWR curve is satisfactory, the design would be competitive with the N3HBX and show less sensitivity to construction variables.

4-Elements and Phased Drivers

We may also use a log-cell driver system consisting of 2 fed element connected by a reversed transmission line. Although related to LPDA structures, small log cells of 2 to 3 elements usually emerge from trial-and-error experimentation when used in conjunction with parasitic elements to form log-cell Yagis. Such cells are normally used for monoband Yagis, but may also appear in some multi-band Yagi designs. A log cell--even if only 2 elements long--generally lengthens the overall length of a Yagi without adding to its overall gain or front-to-back potential. However, the cell is capable of widening the effective bandwidth of virtually all of the Yagi's properties.

For our experimental 40-meter designs using the prescribed element-diameter schedule, I created a 2-element cell with a reflector and a director to form a 4-element beam. The log cell has a normal form, with a longer rear element and a shorter forward element connected by a 250-Ohm phase line using a single half twist. The forward element center serves as the connection point for the main (50-Ohm) feedline. The following table provides the critical dimensions.

Dimensions of the 4-Element Wide-Band Yagis with Phased Drivers

Note: All dimensions are in inches. Multiply by 2.54 for centimeters,
0.0254 for meters.

Antenna Element Spacing Half-Length Tip Length
Short Reflector 0 462 76
Rear Driver 194 439 53
Forward Driver 290 412 26
Director 2 427 399 13
Long Reflector 0 463 77
Rear Driver 192 432 46
Forward Driver 290 401 15
Director 2 467 388 2

Between the two versions, numerous dimensions change by greater or lesser amounts. However, the reflector-to-driver spacing remains almost unchanged, with the major boomlength change between the forward driver and the director. The shorter version is just over 35.5', while the longer version is just under 39'.

As the following table reveals, the shorter version of the antenna achieves virtually all of the array goals except one: it falls just under the desired 20-dB front-to-back ratio at the band edges. Fig. 10 shows the curves and patterns that permit a fuller evaluation of the antenna's potential performance/

Frequency       Free-Space       Front-Back        Feedpoint Z               50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 7.01 18.41 36.5 + j 7.0 1.42
7.15 7.07 21.78 47.1 + j 5.8 1.14
7.3 7.26 19.83 40.9 - j 14.5 1.46

Like the other models that we have surveyed, the performance across the band is very smooth, with only operationally undetectable changes in the radiation pattern from 7.0 to 7.3 MHz. The chief merit in the phased driver system is the achievement of the more rigorous 50-Ohm SWR standard in a beam under 36' long.

The longer version of the phased-driver Yagi extends the boom by 40", making a few changes in the element lengths in order to re-center the performance curves. Since the SWR curve of the shorter design was satisfactory, the reflector-to-rear-driver spacing remained almost unchanged. The additional boomlength or driver-to-director spacing gave the array an unnoticeable increment of additional gain. More importantly, it raised the overall level of the front-to-back ratio by an average of over 5 dB. In the process, the SWR curve improved, mostly due to a narrowing of the total range of reactance change across the band. The following table samples the values, followed by Fig. 11, which presents the data in graphical form.

Frequency       Free-Space       Front-Back        Feedpoint Z               50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 7.10 22.34 40.4 - j 1.7 1.24
7.15 7.15 31.49 44.4 - j 1.6 1.13
7.3 7.35 25.35 43.8 - j 9.9 1.28

The longer phased-driver Yagi's SWR curve does not exceed 1.3:1 across the band and hence rivals the SWR performance of the long-boom (51') 4-element Yagi described in Part 1 (but without the 8+ dBi forward gain). It also provides a detectable improvement in the front-to-back ratio. Even though the cited figure is for the 180-degree ratio, the rearward performance generally holds good for the whole of the rear quadrants.

Fig. 12 compares the outlines of the short and long Yagis with phased drivers. The increased spacing from the driver to the director is clear in the long version sketch.

Optibeam of Germany makes available a 4-element Yagi with a phased driver pair. The boomlength is similar to the one used in this design. I do not know the element taper schedule or precise dimensions used by the commercial offering. It does use some inductive loading to shorten the elements. This fact should not alter the SWR curve or the front-to-back ratio in any noticeable way. However, it may reduce forward gain slightly, depending upon the actual degree of element shortening and the inductor Q. These notes are neither for or against the Optibeam design. Rather I note them in passing to confirm that the phased driver system is a viable technique for obtain relatively high performance across the entirety of the (U.S) 40-meter band.

3-Elements and Phased Drivers

Let's review a couple of facts about 40-meter beams as presently used on the band. First, the most common varieties use boomlengths of about 20-22 feet, most of the time with only 2 elements. The wide-band design in Part 1 calls for 2 elements on a 23.5' boom. For reference, let's repeat the dimensions and the performance table.

Dimensions of the 2-Element Wide-Band Yagi

Note: All dimensions are in inches. Multiply by 2.54 for centimeters,
0.0254 for meters.

Antenna Element Spacing Half-Length Tip Length
2-Element Reflector 0 451 65
Driver 282 412 26


Frequency Free-Space Front-Back Feedpoint Z 50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 6.50 9.89 40.7 - j 19.4 1.60
7.15 5.96 10.70 53.1 + j 3.5 1.09
7.3 5.52 9.71 63.5 + j 24.2 1.63

The gain and front-to-back values are typical for a 2-element driver-reflector Yagi. We obtained full-band coverage by widening the element spacing from the usual 40-meter 20' value.

The second fact from Part 1 is the ability of Yagis to operate satisfactorily without a reflector, if the driver cell can handle the chore of setting the basic feedpoint impedance level. In the phased driver cell used in the preceding pair of designs, we have such a cell. If we remove the reflector, the impedance does not change beyond our ability to adjust by lengthening the director by a few inches on each end. We may leave all other elements just as they occur in the 4-element design and also retain the spacing values. The result is a 3-element Yagi with a phased pair of drivers and a single director on a 22.9' boom. Fig. 13 shows the outline.

The following table and Fig. 14 surveys the potential performance of the shorter array across 40 meters.

Dimensions of the 3-Element Wide-Band Yagi with Phased Drivers

Note: All dimensions are in inches. Multiply by 2.54 for centimeters,
0.0254 for meters.

Antenna Element Spacing Half-Length Tip Length
3-Element Rear Driver 0 432 46
Forward Driver 98 401 15
Director 275 393 7

Frequency Free-Space Front-Back Feedpoint Z 50-Ohm
MHz Gain dBi Ratio dB R +/- jX Ohms SWR
7.0 6.13 13.56 38.2 + j 5.5 1.35
7.15 6.57 15.30 48.4 + j 1.8 1.05
7.3 7.05 13.49 45.2 - j 15.2 1.40

The data suggest a Yagi with performance intermediate between a standard 2-element driver-reflector design and the full 4-element phased-driver design with a reflector. Indeed, the performance is similar to that of an optimized driver-director 2-element Yagi, but without the usual restriction that comes with such designs--an exceptionally narrow operating bandwidth. Both the gain and front-to-back ratio exceed what is possible in the 2-element Yagi, but do not quite reach the level achieved by the 4-element design. However, the patterns are very well-behaved, and the 50-Ohm SWR curve never rises to 1.5:1.

Adding the weight of a third element to a 2-element Yagi is not a small adjustment. The overall beam will weigh about 30% more and call for support strengthening. However, the shorter boomlength reduces overall structural stressing in the wind relative to one of the longer boom 4-element designs. With a 23' boom, the 3-element phased-driver arrangement may be the most promising way to achieve the desired contester SWR curve with quite reasonable performance everywhere in the band. Indeed, the short boom allows room for further tweaking of the phased drivers to allow even further improvements in the SWR curve, as well as further adjustments to the length and position of the director for small performance improvements. The design improves on the sample 12-meter antenna (with a loss in gain and front-to-back ratio) by beginning with a wide-band phased pair of elements rather than the maximum gain pair used in the earlier experiment. In addition, the design also shows what the reflector in the 4-element long design adds to performance--assuming that its 39' boomlength is not a hindrance to implementation.


These designs complete our small survey of wide-band 40-meter Yagis. They do suggest that there are a number of alternative designs capable of covering the band with adequately high performance with boomlengths under 40', with not much lost as we shorten the array down to 35' or so. The new additions appear to have the additional merit of not being as sensitive to changes in some critical elements as the N3HBX design. Indeed, the final 3-element Yagi with phased drivers seems to be the most practical among all of the designs

Along the way, we have had occasion to probe various Yagi properties often overlooked by both Yagi users and home Yagi builders. Since few radio amateurs have the wherewithal to build or sustain a full size 3- or 4-element 40-meter Yagi, these additional discussions may prove in the long run to have more utility than the design suggestions. Indeed, the designs would require considerable revision and refinement to meet whatever one chose as a usable element taper schedule for implementation. Hence, they are only first-order design examples useful for comparisons among the collection of Yagi ideas.

The U.S. 40-meter amateur band is a special case within the HF spectrum. It is just narrow enough to lead us to believe that we can cover the entire band with a single antenna. However, the mechanical requirements for tubular elements restrict the achievable bandwidth with standard designs. Full band coverage calls for design ingenuity and some specialized techniques for extending coverage while maintaining a manageable boomlength (assuming that element length presents no insurmountable problems). As a consequence, it is a good band for learning one or two new things about wide-band Yagi performance and design.

Go to Index