|
|
CQTFx. tbl. 200x |
Efni þessa blaðs |
Mobile VHF AntennasBy Gudmundur Loeve TF3GL IntroductionThis article contains various investigations of mobile antennas for VHF operation. Since this is an on-line publication, it may be updated or amended at any time, and will grow with the author's knowledge. Antenna modelingEZNEC model descriptionEZNEC+5.0 was used to model a Nissan Patrol 4WD, using realistic dimensions, materials (steel, 7×10 -7 ohm×m) and wire thickness when modeling surfaces, as recommended in NEC2 documentation. The car model contains roughly 1300 wire segments, has all outer surfaces (including a bottom surface for ground capacitance) and is placed 0.3 m over "bad" ground (G = 0,001 S/m ; k = 4). Model considerationsAll the antennas being tested are attached to the car body at wire junctions to minimize calculation artifacts and inaccuracies. Only one (resonant) antenna must be attached at a time, to avoid additional resonances. Due to the 1500 segment limitation imposed on EZNEC+, the wire length of 0.3 m used for most surfaces in the model is slightly long for the 2 m bands. This is countered by selectively adding segments to the model's top surfaces (hood and roof), that are in closest proximity to the antennas and carry the most current. A future improvement of the model could be to re-draw it in its entirety using 0.2 m long segments, but this would not fit into the 1500 segment limit of EZNEC+. It is my opinion, however, that for the intents and purposes of these investigations, this does not significantly alter the results. 1 However, it may be that the exact "ups and downs" in the far-field intensity plots should be taken with a grain of salt – which is a moot consideration anyway in a mobile installation where the ground is neither completely flat nor is the vehicle stationary. The antennasTwo antenna types were modeled: a 1/4 wl. antenna and a 5/8 wl. antenna. Both are modelled with 3 mm thick steel, the former without any kind of loading, but the latter including a ~700 ohm inductive reactance (loading coil) and a modest 1 ohm loss resistance (coil loss), considering the very high-Q construction of the typical 5/8 air-wound, self-supporting coil. Two types of antenna placement were tried: On the rear top of the car and on the left side of the hood. Both antennas were tried on the top, but only the longer 5/8 wl. was tried on the hood. ConclusionsWhile the differences between the humble quarter-wave and a full 5/8 wavelength antenna when placed on the top of the car roof aren't that great (around 2 dB in favour of the 5/8), placing the 5/8 on the bonnet (hood) is decisively more flaky: There are nulls around 6 dB down (quarter power) from the average due to shielding by the car body (although the shielding effect is probably more uniform in the real world than in the model, which is a bit coarse-gridded for VHF, as mentioned before). It is doubtful that the added inconvenience of a 5/8 is worth the effort except in a few very marginal situations, as it has only a very slight gain improvement over the 1/4 wave, while the 1/4 wave is both more broadband and less obtrusive.
Notes
|
Far-field performanceOf primary interest were the absolute gain figures at low angles of the different antennas, mounted in different places on the car. All simulations were conducted at 145 MHz with resonant antennas. For convenience, and because skywave is of no interest for mobile VHF, all the results are plotted into the same eleavation-type graph. Two elevation slices are taken: at 2° and 10°:
|
