160 meters - or bust!
We start with my classic suburban lot (100' x 70') outside Boston which provides little room to build antennas for the lower bands.
A 100-foot dipole @25 feet above the ground (along the back property line) tuned well enough (through an MFJ tuner) for 80 meter WAS, performance on 160 meters was sub-par. It's easy to see why, since 25 feet is a fraction of a wavelength on 160 - meaning most of my signal was going up, not out, as we see in this EZNEC plot:
After exploring options including a sloper, inverted "U," several variations of an "L," some home-brew verticals with capacitive "top hats," as well as verticals sold by DxEngineering and MFJ, I decided to try an experiment suggested by Woody, WW1WW, to convert my 100-foot dipole, which was fed by twin-lead, into a vertical.
The EZNEC plot shows, with the vertical, I would achieve the goal of radiating most of the station's RF at an angle:
Radials, essential for a establishing a RF current return path to ground, would need to be installed. There were three challenges: 1. The small footprint of my property. 2. My house sits only ten feet from one side of my lot. 3. The only tall trees were also along the back end of the lot. So laying out radials would require some creativity. I began with four long radials at ground level:
It didn't work very well. Even with my MFJ tuner a decent SWR was impossible. To tune up a vertical only 25 feet high on 160 was going to require a loading coil.
Using EZNEC Woody modeled the antenna and calculated parameters for a base-loaded coil; 38 turns of 10-gauge wire coiled 2" to electrically extend the antenna. It would also eliminate the need for the balun. I constructed a coil using a short piece of PVC. A hole was drilled at one end for a binding post with banana jack (I was seeking an easy way to attach the radials.) An SO-239 was also installed.
The coil is connected to the center of the SO-239 on the inside of the tubeIt exits from the top of the pipe through a plastic cap. The cap and all holes were originally sealed using my wife's hot glue gun (a QST "Hints and Kinks") but, frankly, the glue did not provide a waterproof or lasting seal, so I had to redo with caulk. Now let's look at the performance of the vertical with the coil. I've included a brief reminder of what each measurement means...
This is the ratio of the antenna impedance (Z) to the transmission line characteristic impedance (50 Ω). (My buddy Doug, K1DG, says it's basically a measure of how happy the transmitter is.)
This is reactance, the opposition to alternating current due to a combination of capacitance and/or inductance (present in every antenna.) An antenna is resonant when capacitance and inductance cancel each other out.
This is radiation resistance, which measures the loss from the system (antenna, conductors, radial system). In this case, it is a reliable measure of how well the radials are performing as a counterpoise - an RF path to ground.
SWR and X were very good but R was over 50 ohms across the entire band. The four radials were a poor counterpoise and performance would suffer. More radials were clearly going to be needed, but you've seen the size of my property. Where could I lay them?
The answer was found in the work done by Rudy, N6LF in several QEX articles on verticals and radials which detail how elevated radials can be as effective - and sometime more effective, than ground-based radials. I ran a pair of radials approximately 3 1/2" off the ground, along the upper support of my fence:
These three plots show the original four radials (in red) and the two elevated radials (in blue). NOTE THE MFJ ANTENNA TUNER WAS NOT IN CIRCUIT. The SWR rose a bit but was still under 1.5:1 at the target frequency of 1.840 MHz.
Reactance was also lower at our target frequency.
What's most important is that R dropped almost 20 ohms across the band. An unqualified success!
Over the next month I experimented with several layouts of radials, including two on the top of the fence and another test with a "spider" of eight 16' radials (testing the theory that more radials, even short ones, will improve antenna performance.) None lowered R. In fact (and I challenge you to do the math - because I can't) both tests actually raised R across the entire band! In late August, 2011 I decided to try a couple of longer, ground-level radials.
The results, plotted in green, show a slightly higher SWR
Reactance did not change appreciably from the levels measured with 4 (red) or 6 (blue) radials.
The good news was R. The extra 2 radials reduced R across the entire 160 meter band by another 5 ohms, as low as the 30 ohm range around 1840 (the FT8 frequency)