20 meter phased beam 1, Casey 0

That shortened phased beam that you've been reading too much of lately? Finally whupped me! I was close, but no cigar. In the ultimate last-ditch push I jettisoned the Guanella balun, tried different lengths of phasing line, and I felt I was getting close, but resonance was still out of reach, at least without a bagful of reactance and that just wouldn't do. I was able to tune out the reactance with a parallel capacitive stub using some scrap coax, but 4nec2 told me that I had to kiss about 2dB of gain goodbye if I did that.

So, it was time to do some head scratching and RF soul searching. What could I salvage? I removed the phasing line and then did what I should have done in the first place - 20/20 hindsight being the better part of valor -, which was to utilize the neglected 4nec2 Optimizer!

The Optimizer is a really fun sub-program in 4nec2 that, once your design is close, lets you select various aspects of your design (usually wire length), weight the characteristics most important to you (e.g., SWR, Gain, F/B, ...) and then you let it rip. It chugs away for several to many iterations after which you can accept the changes or tell it to keep going. Once it's done it's a one-button click to add its changes to your NEC file. I didn't use it on the phased design as I thought I'd probably tried every permutation already and I was getting sick of that design anyway.

Thus, I put the optimizer to work on the original KJ5VW design. I'd already found manually that moving the position of the coils or changing their values was more or less counterproductive so I left those parameters be. I was also trying to overcome a limitation of the original design, which was that the Three Sirens of antenna design, low SWR, high gain, and high F/B didn't seem to line up well at any frequency.

Well, the optimizer did the trick, increased the gain and lined it up pretty well with F/B and reasonable SWR (though not the lowest). I end up with a nice 25 degree TO angle, over 11 dB gain and narrow bandwidth along with not too shabby F/B of 15 dB. I can live with that! 

After taking these data down to the shack and tuning up 17 meters on a now much more reasonably behaved antenna, I set to work on the 20 meter tails. Overall, on 17 meters the antenna is about 13.5 feet wide (4.11 meters) and 16.75 feet wide on 20 meters (5.1 meters). Other than the specific element lengths the only notable difference between this design and the original is that the reflector is about 3 to 4 inches longer, whereas the elements are equal in the original.  Also, the separation of the elements is 2 feet closer in my version.

You can see how I finally cured the element droop, by adding a short brace to each end to which I could strap the element support poles. 

 Electrically, after removing the capacitance stub and Guanella balun I only had to add a jumper to connect the reflector elements.
I also strengthened the center support that slips over the mast (not pictured). 

The weight  I don't know exactly but I'd be very surprised if it was even close to 10 lbs., so it's very easy to haul around if there are no obstructions. Breakdown and re-assembly should not be difficult, but I'll know more once I set this up on an outing. 

It hasn't yet been put to an authentic field test as the place on the property where I constructed it is in an RF "hole", so even though I was able to A/B it with the quad, the signals they were both receiving were coming in at a high angle, so real F/B remains to be seen. After I make some deferred mods to my mast up by the house I will mount it there with a rotator to see how it performs in the clear. 

By the way, starting with this antenna I will put together a ZIP file of pictures, 4nec2 screens, and the NEC file on my Antennas Page instead of just the NEC file. 


No comments:

Post a Comment

Thanks very much for your comment! 73, Casey