Currently I'm using only one antenna system. It is an 80 m full size vertical with about 30+ radials. The vertical is fed via 9,0 m of RG-213 coax attached to antenna tuner in the hamshack. For operation on other bands I'm force-tuning the system. Obviously the coax may radiate as well ;o)
18 m SpiderPole vertical
The vertical is based on 18m long telescoping fiberglass pole called SpiderPole (http://www.spiderbeam.com). I ordered it from Germany over Internet. Within a week a 7 kg package was delivered to my door.
The vertical is fastened to my house which has metal roof. The placement is because of lack of space on the plot. The lower mount point is about 2,5 meters from ground level and the tip of the pole is at about 20,5 m height. A 2mm OD enameled copper wire was inserted inside the pole.
I have dug about 30 radials (variying from 4 to 20 m in length) 2-3 cm into the ground using bare 2,5 sq mm household wire (stripped the insulation off). Currently I have about 600 m of wire in the radial system. The house and nearby lot border restrict layout of the radials. They cover only about 2/3 of full circle in a bent hourglass shape.
Spiderpole is very flexible. Initially it was guyed at two levels. After a couple of high wind storms the lower guy attachment had a bolt twisted and the attachment broke off. The pole had been able to flex between the gying points in the gusts. I added one more guying point and now the pole has been very stably upright even with 25+m/s winds (nearly 60mph). The pole is guyed at 10, 13 m and 16 m leaving a 4,5m top whip free. As the pole is lightweight I chose to use 3mm Dyneema yachting rope for the guy wires. The rope can easily handle several 100 kg tension. The ropes will stretch a couple of per cent within the first weeks. So I used yachting fasteners for easy tightening. I also have two 2mm Spectra ropes with pulleys attached at the 16m guying point for future experimenting with wire antennas.The pole has been up since October 2007 and sustained cold Nordic Winters, too, without any problems. The antenna system seems to be working fairly well - 240+ countries /all zones worked so far. Well, 500 W amplifier helps a bit.
I have started to model the vertical with 4Nec2 -antenna modeling software. However initially I got mixed results compared to the measured ones. After setting up the vertical, I measured it with AIM4170 vector network analyser. To my surprise, the resonance point was ca 7% lower than anticipated from the length of the vertical. I thought that the proximity of the house/metal roof required some closer study in the modelling. Now the resonance lies at 3.27MHz instead of 3.5MHz of the model. However the vertical works. Obviously I need to tune it shorter.
Currently the feed point impedance at resonance is about 52 ohms. About a year ago I had 10 radials and the impedance was about 72 ohms. This shows the effect of adding radials even though they do not cover full circle. Discussion on eham.net's Towertalk forum revealed that the pole's dielectric may be the culprit for the downshift of the resonance. Looks like the vicinity of the house was not a major contributing factor on the model discrepancy. Fiberglass has dielectric factor of some 4.0-4.9. The pole wall thickness is 1 mm at the top and about 2 mm at the bottom. Since the element wire runs inside the pole, the pole acts as a dielectric coating on the wire. Using 1,5mm thick coating on the wire the 4Nec2 model resonance shifts down to 3.3 MHz, pretty close to the observed. However the model feed point impedance is 40 ohms compared to measured 52. Measured and modelled bandwiths for SWR less than 2 are about the same, 260 kHz.
I have scanned the full HF spectrum, too. The model gives resonance points for the 2nd, 3rd and 4th harmonics quite close to observed. Above that they differ more, but that is not so relevant. The measurement reveals bumps in phase angle near 4.7 MHz and phase flattening in the 18-19 MHz area. These may be caused by the house/metal roof.