Small Packrat Logo

Using "Yagi Optimizer" to Design the New W3CCX 6M Antenna

by

Gary Dallas, WA1YHO

 

For the past few years, the W3CCX June contest effort has used a Cushcraft 617-B6 Boomer antenna on 6M. While it's a fine antenna, the front to rear performance was not what I would have liked. Our contest site in the Pocono Mountains is surrounded by a lot of other contest stations and the QRM can be fierce. Having just a few more dB of rejection off the back of the antenna would be a big help.

I've also wanted to try stacking a pair of antennas on 6M. My experience on other bands is that stacked antennas seem to "play" better than the stacking gain would imply. For practical reasons, this meant that the new antenna would have to be smaller than the 34 foot boom Cushcraft. I set a goal of designing an antenna on a 20 foot boom with a 30 dB front to rear ratio. The following paragraphs detail the design and construction of the new antenna.

DESIGN

I had been playing with the yagi analyzer program called YA that is included with the recent ARRL Antenna book. It will compute and display antenna patterns for yagi antennas. By moving element spacing and lengths, you can see the effect on the beam pattern. However, this manual process of moving an element and manually noting the results was very tedious.

YA is really just a stripped down version of YO, Brian Beezley's (K6STI), Yagi Optimizer program. (Available for $60 from Brian, see ads in QST.) YO will automatically optimize a yagi design based on the weighting of three parameters: forward gain, SWR, and front to rear ratio. You chose the weightings you desire. In my case, I was willing to sacrifice forward gain to improve the front to rear ratio. After you set the optimization tradeoffs, the program will reposition and modify element lengths to optimize for the desired performance. No more tedious cut and try! The computer does the work.

Brian includes quite a number of HF and VHF antenna designs with his software and I was pleased to find that the Cushcraft 6M Boomer was there. This was important, as he had already done the work to compensate element dimensions for the boom mounting method. He did this by shortening the inner element section length by 0.4 inches for the purposes of simulation. So starting with the Boomer design, I tried a number of variations. First, I tried using all six elements. The optimized designs always had too long a boom. Next, I just deleted the 4th director from the Boomer element list and ran the optimizer.

The optimizer came up with a five element design on a 21' 3" boom. From the Figures 1 and 2, it can be seen that the forward gain is only about 0.6 dB less than the much larger Cushcraft and that the F/R is 10 dB better! According to YO's figure of merit number, emphasizing the F/R performance cost about 0.6 dB of forward gain over what could obtained with the 21' boom length. Figures 1 and 2 are screen dumps of the main YO screen. The numbers inside each beam pattern are, from top to bottom, frequency, forward gain, front to rear ratio, input impedance, matched standing wave ratio, and figure of merit. The beam patterns are free space E-plane plots at the specified frequency. Figure 3 overlays the two antenna's free space E-plane beam plots at 50.150 MHz. The four graphs in Figure 4 plot the gain, F/R, SWR, and impedance verses frequency for the new design.

CONSTRUCTION

One of the nice things about the 6M Boomer mechanical design is that it lends itself to modification. The elements are in three pieces. A 48" long section of 0.750" tubing forms the center section of the element. A piece of 0.675" tubing slides into each end of the center section and is clamped with a stainless steel hose clamp. The elements mount to the boom with U-bolts and saddles. The boom is made up of sections of decreasing diameter tubing telescoped together and held with hose clamps.

The Cushcraft parts were usable in the new design with one close dimension. The new first director was quite a bit longer than the stock Boomer design. However, there was just enough of the outer element section to clamp into the inner section. Good enough for a weekend contest effort, but I might get a longer piece of element material for a permanent installation.

For the temporary contest installation, I didn't use the two outer boom sections and telescoped one of the inner sections in a bit to obtain the desired boom length. For a permanent installation, I would nest all of the boom sections for added strength. The stock 34' boom antenna needed boom supports, but, at 21', the new antenna did not need them.

I knew that the match would be different from the stock antenna given the impedance numbers reported by YO. However, I was sure that I could tune the driven element using the Cushcraft T-match hardware. I had to lengthen the driven element a bit and move the T-match shorting bars, but a perfect match was obtained.

The final element dimensions and spacing are listed below. The 'outer section length' listed below measures the distance from the end of the inner element section to the end of the element.

Element
Spacing
Outer Section Length
Total Element Length
Reflector
0.0"
35.095"
118.19"
Driven Element
50.17"
33.125"
114.25"
1st Director
92.89"
30.066"
108.13"
2nd Director
188.53"
28.816"
105.63"
3rd Director
254.58"
27.600"
103.20"

The details on the T-match section are as follows: Set the shorting bar 11.125" in from the end of the T-match bar.

In summary, the new antenna was easy to build using the stock Cushcraft Boomer parts.

ON-AIR RESULTS

Of course, computer simulations are, well, just simulations. The real test is on the air. And the antennas worked great! QRM was greatly reduced. Much of our local QRM comes from stations east and northeast of our contest location. When operating in the southwest, as we did a lot of this year, these stations were just not the problem they used to be. And the antennas seemed to work well. The pattern was as clean as the simulation showed and, while it's hard to make measurements with on-air signals, the S-meter said the F/R was close to the predicted 30 dB.

In my opinion, YO is a must have piece of software for anyone who wants to build or modify antennas. While there are many subtle problems in modeling antennas that I did not discuss (and probably a lot more I don't even know about yet), I am convinced that whether you're looking for that last few tenths of a dB of forward gain or a better pattern, this program will help you design a better antenna.

THOUGHTS on STACKING 6M ANTENNAS

Unfortunately, contest time came up quickly this year and I didn't have time to study how best to stack two of these new antennas before the contest. Based on the available 20 foot mast and the mounting position of the rotor in the tower, they could be stacked a maximum of about 17' apart. Lacking any scientific insight at contest time, they ended up about 16' apart. See picture below of W3CCX/3 6 meter antennas.

However, while preparing this article, I did some further investigation. First, I dug back through my VHF reflector archives and discovered a couple of very informative posts from Zaba, OH1ZZA. He recommends a 0.6 wavelength spacing for stacking 6M antennas, especially for antennas with short booms and poor patterns. He states that the 0.6 stacking distance greatly reduces sidelobes in all directions which, in the H-plane, can improve cosmic noise rejection.

I checked this using YO to plot the H-plane free space response of a pair of vertically stacked antennas. Sure enough, there is a large dip in the H-plane sidelobe response at a 0.51 wavelength stacking distance for this antenna. In fact, in free space, the H-plane lobes all but disappear!

Seventeen feet is about 0.84 wavelength. The sidelobes at this spacing were much higher and the gain was only 1 dB more than the 0.51 wavelength stack. Next year I think I'll try a closer stacking distance, something around 10 feet. I think I hear the weak ones better already!