W5ALT Home


Electromagnetic Waves
Basic Concepts I
Basic Concepts II
Antenna Scaling
RF Ground
Antenna Tuners

Horizontal Dipoles
Vertical Dipoles
Inverted Vee Dipoles
Bent Dipoles
Short Dipoles

Ground Plane Verticals
1/4 Wave GP Verticals
Ground vs. Radials
Short Verticals

Practical Stuff
Case Study 1
Case Study 2
Case Study 3



Restricted Space Antennas

by Walt Fair, Jr., W5ALT

Short Verticals

In the previous section we saw that the ground plane vertical can be an effective antenna and can be used in a space limited situation, since the vertical takes up little area. However, it may still be too big for practical reasons due to the need for radials. In addition there may be height limitations that cause problems. Especially if we are forced to use indoor antennas, getting a full 1/4 wavelength in a living room may be impractical. In this section we'll look at how to further shorten the vertical part of the antenna and what effect that has on performance. Later we'll look at how to keep the radials down to a manageable size in restricted situations. And finally we'll look at what happens when both the vertical element and the radials have to be shortened.

Shortening the Vertical. First, let's look at a ground plane vertical for 80m (3.6 MHz), which as in the case of a shortened dipole, is a big challenge for limited space antennas. In this case we'll assume that the radials are the right length (about 68 ft long) and the base of the antenna is 10 ft above an average ground.

As can be seen in the left hand figure, as the vertical is shortened, the resistive impedance drops from about 36 ohms to about 1.5 ohms. Meanwhile the capacitive reactance increases from zero at resonance (68 ft) to more than 5000 ohms at very short lengths. Just as we noticed for the short dipoles, this has severe implications when we consider that any matching coil will have resistance due to the coil Q which will be greater than the resistive impedance at short lengths. But, it gets worse!

As the right hand graph shows, the take off angle doesn't change much, but stays around 20 - 25 degrees. However, the gain changes dramatically, going from near 0 dBi (equal to an isotropic antenna) to 10 and even 15 dB of loss at small vertical lengths. In effect, as the vertical element is shortened, the radials become relatively more important and their interaction with the ground dominates the antenna performance. Hence most of the signal is lost in warming the ground (10 dB loss represents a 90% loss in radiated power). So the net effect is that with small vertical elements, we will lose a lot of power in the matching network, then most of the rest will simply warm the ground. That certainly isn't very encouraging!

Shortening the Radials. Now let's look at what happens when we have plenty of vertical space, but no room for radials. Once again, we'll look at an 80m vertical at 3.6 MHz, but now assume that the vertical element is the right size, about 68 ft. Once again, the base of the antenna will be 10 ft above an average ground.

As the left hand graph above shows, the resistive impedance stays in the range of 35 to 40 ohms for nearly all radial lengths. Meanwhile, as the radials are shortened, the capacitive impedance rises but doesn't get extremely large until the radials become very short, less than 10 ft. Both of those observations are encouraging, since we can always add a simple coil to cancel the reactance and be left with a resistive impedance that will work well into 50 ohm coaxial cable.

The right hand graph is also interesting. As can be seen, the take off angle stays at 22 - 23 degrees, no matter what the length of the radials are. In addition, the gain stays steady, too, at just below 0 dBi. Why is this so much different than the previous case where the vertical was shortened?

If we compare the geometry of the two situations, in the first case it is apparent that the radials dominate the antenna structure when the vertical element is short. This causes most of the radiation to penetrate the ground, with the associated losses. We could say the antenna is mostly cooking earth worms. When the radials are shortened, however, there is less radiation penetrating the ground, so less losses occur. The impedance is still adversely affected, but at least we are not heating the ground as much. This clearly shows that if we have to shorten either the vertical or the radials, it is much better to shorten the radials!

Shortening Everything. Unfortunately sometimes we don't have room for either a full size vertical element or full sized radials. In this section we'll see what happens when we have to shorten everything. Again, we'll consider an 80m antenna at 3.6 MHz with its base 10 ft above an average ground. In the shrunken model, both the vertical and radial elements will be shortened, but their sizes will be kept equal.

As shown in the above graphs, the performance is similar to the case where only the vertical element is shortened, but not quite as severe until the lengths are very small. The gain does not drop quite as rapidly and the take off angle changes slowly. The resistive impedance, however, drops quickly and the capacitive reactance rises as the antenna is shrunken.

Essentially, as the antenna shrinks, it changes impedance since the size is too small to be anywhere near resonance. However, since the relative effect of the radials is not decreasing, a substantial part of the energy is lost in a combination of resistive wire losses and ground losses. In trying to feed this antenna when it is small, much of the transmitter power would likely be lost in the resisitance of any coil used in matching, just as described above. Although not as bad as keeping the radials long, this isn't a very efficient antenna when it gets very short.

Since we saw a significantly better performance when the radials were shortened, it might seem possible to try to keep the radials shorter than the vertical element. The idea in doing this would be to minimize the interaction between the radials and ground, while still shortening the vertical element. Unfortunately, although not presented here, the results are not very different from the case where the radials are the same size as the vertical element. It appears that there is little remedy for the problem. If you can keep the vertical element as close to full size as possible, you will be much better off.

Effect of Diameter. We showed in the section on dipoles that a larger conductor diameter was better and that below AWG #20 wire, the resistive losses in the wire seemed to become important. So far, all of the antennas in this section were modeled using #14 wire. What happens if we use larger conductors? To evaluate this, we use a 15 ft vertical element with a varying diameter while keeping the radials 15 ft long, but made with #14 wire. The base will still be at 10 ft above an average ground.

As the left hand plot shows, when the wire diameter is below about 0.1 in, losses (indicated by the increased resistance) start to get more important and are worse for aluminum than for copper conductors. However, if the diameter stays above about 0.1 in, the effect is minimal. Obviously, since every little bit helps, using as large a diameter as possible is best, but especially above 1 or 2 inches, the effect isn't important. In addition, for diameters larger than about 0.5 - 1 inch, the difference between aluminum and copper conductors is negligable. Also interesting is that there is no difference between copper and aluminum is terms of the reactive (capacitive) impedance and only the resistance is affected by the type of metal.

The right hand graph shows the gain as a function of the conductor size. (The take off angle varies very slightly, always around 26 - 28 degrees, so is not shown). As can be seen, copper conductors always give more gain, but for diameters larger than about 0.5 in, the difference between copper and aluminum is small. Generally the larger the diameter, the more gain that will be obtained, but as the diameter gets larger, the benefit is less. It appears that from gain, impedance and structural considerations, 1 - 2 inches is perfectly acceptable and further increases in diameter probably aren't worth the extra weight and cost of metal. After all the total diference between 0.1 and 10 inch diameters is less than 1 dB, which probably wouldn't be noticed.

As a result of looking at various means for shortening a vertical antenna, it seems apparent that we want to keep the vertical element full sized if at all possible. However, when that is not possible, shrinking the antenna can be expected to cause low resistive impedance, high capacitive reactance and problems with low gain due to resistive losses and ground losses. We also have seen that increasing the diameter of the conductor helps minimize losses, but that going to more than 1 or 2 inches diameter isn't usually worth the cost and effort.