## Mobileantenna radiation patterns are seldom circular their shapes depend on where the antennas are mounted

Sometimes moving your parked car just a little allows you to make a 2 meter contact that you could not make before—or breaks one you had going. I've experienced that many times—even out in the country away from buildings, but I did not realize until I completed this study, that turning the car to face a different direction could make an even greater difference.

The Reason

Vertical antennas on cars rarely have circular radiation patterns. Their patterns depend on their loca-

2493 Malvern St Cambria, CA 93428 e-mail [email protected] tion on the car and on the shape and construction of the car.

### Car Roofs and Trunks are not Symmetrical

Feeding current into the base of a vertical antenna mounted on a car forces an equal and opposite current to flow along the car body, radially, away from the antenna. At 2 meters, the radial path in the car body is typically more than 7JA long in some, if not all, directions thus creating rings of standing waves of current and voltage at various distances around the antenna. Voltage nodes form where these radial currents stop at the far extremities, which (in most directions) are at the lower edges of the car body, down near the ground. Current nodes form UA back towards the antenna. Other voltage and current node "rings" form in complex shapes at X./4 intervals back to the base of the antenna. The sides, front and rear panels of the car tend to act like a very fat, nonuni-formly driven, k/4 vertical with a voltage node at the lower edge near the ground and a current node 20 inches higher, near the upper edge of the body. The total far-field antenna pattern results from the combined radiation from all these horizontal and vertical patches or "rings" of current and voltage nodes.

From one point of view, the location of the "actual" antenna on the car merely serves to determine the patterns and relative current densities of a complicated set of standing waves. To a large degree the electromagnetic radiation emanates from the entire structure—not just from the piece of metal we call the "antenna."

Essentially, the same thing happens in reverse during reception. The received fields induce currents and voltages along complex paths all over the conductive surfaces of the car. Local reradiated fields from these couple to the actual antenna to generate the resulting current in the feed line.

The car acts as a contoured ground-plane with a three-dimensional resonant response.

### The Model

I used a simple U4 whip for this study and modeled it in the four locations shown in Fig 1: at the roof center (point A), at the center of the front roof edge (B), at the left-front corner of the roof (C) and at the center of the trunk lid (D).

The "typical generic" car model is loosely based on a 1987 Ford Taurus with most dimensions rounded off to the nearest eight-inch increment for modeling convenience, as shown on

The conductive surfaces of the car are modeled as a grid of 678 wires, most of which are eight inches long (0.1 X) and all of which are divided into segments no longer than four inches (0.05 X). The model contains a total of 1456 such segments. Some long members, such as the window posts, contain many segments and act as dis tributed inductors connecting the various surfaces, just as they do in reality. Experience and analyses show that we could improve the accuracy of this model by using an even greater number of shorter single-segment wires, but the results would not change materially. The computer run times would be even longer—life is a series of compromises.

I used EZNEC-M1 to perform the modeling calculations. EZNEC-M is an easy-to-use antenna modeling program based onNEC-2, which is a well-proven but cumbersome program. NEC was originally designed for main-

1EZNEC and EZNEC-M, by Roy Lewallen, Fig 1—I've modeled a typical car body as a grid of eight-inch modules. There are W7EL, PO Box 6658, Beaverton, OR four antenna mounting locations (A) center of roof. (B) front edge of roof. (C) 97007. corner of roof. (D) center of trunk lid.

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Elevation Plot through Main Lobe

Azimuth Plot e

Azimuth Plot

Azimuth Plot

Frorjt of Car

Center of Roof

Front Center Roof Edge

Frorjt of Car

Center of Roof

Front Center Roof Edge

Azimuth Plot

Left Front Comer of Roof

Azimuth Plot

Azimuth Plot

Azimuth Plot

Left Front Comer of Roof

Center of Trunk

### Center of Trunk

Fig 2—The lower four plots are azimuth patterns for a X/4 vertical mounted at four places on a car. The upper four plots are elevation patterns through the main lobes. All azimuth patterns are for 0° elevation. Outer rings are at 10 dBi.

0 0