The antenna is the last link in the circuit between the transmitter and the listener. In this link, there may be a phasor (for a directional station) and some antenna tuning units (ATU). In any case there will be at least one ATU. Let's take a look at the function of a typical ATU.
In the case of a nondirectional station, it is necessary to match the output of the transmitter, which is usually 50Ω, to the base operating impedance of the antenna. A typical antenna might have a base impedance of about 40+j100Ω.
Figure 1. An L network is not normally used in a directional array phasor because its adjustment is too coarse.
The link between the transmitter and the antenna is a simple piece of coaxial cable that has a broad bandwidth. However, when connected to an ATU it can become part of a circuit that may introduce unexpected distortion into an AM signal. In reality, it is part of the transformer that matches the antenna impedance to the transmitter output impedance. Optimum transmission requires that the two side bands are equal in amplitude and phase so that the resultant of the two side bands falls directly on top of the carrier phasor. If this condition is met, the received signal will be a close replica of the original.
There are several influences produced by the antenna tuning unit network and the antenna that can affect the bandwidth of this important link. In a vector diagram of the circuit there are three phasors, one for the carrier and one for each side band. If the amplitude or phase of any one of these is not correctly adjusted signal distortion or incidental phase modulation can result.
If one of the side bands is missing or is severely attenuated the signal envelope will not be a sine wave. The degree of distortion depends on the amplitude of the affected side band. In the unlikely event that AM stereo is being transmitted the distortion will be greater and stereo separation will be reduced.
The bandwidth of the transmitted signal will not be excessive. Even with severe distortion signal splatter and adjacent channel interference it would not occur because it is in the receiver that the carrier and side bands are combined and distortion is generated. So distortion may not be perceived on the on air monitor in the studio, but only in listener's receivers.
There are several ways to design the circuit between an AM transmitter and its antenna. Before the days of computers, pocket calculators and Burrough's calculating machines, consulting engineers used slide rules to design antenna systems. Often the first circuit parameters that gave the desired pattern or match were selected, and design work stopped. Now computers make it possible to iterate designs quickly until the best circuit values are found. In the case of nondirectional stations the procedure is relatively simple but for directional antenna systems a great deal of work is still required to ensure that unanticipated problems do not arise.
In a nondirectional AM station the base operating impedance of the antenna is of prime importance. The R part of this impedance multiplied by the antenna operating base current squared defines the operating power. The ±j (reactance) portion is transformed by the ATU to match the 50Ω coaxial cable.
It is not practical, in most cases, to develop an AM antenna with an impedance of R±j0. There are many filter techniques available to affect the desired match. The most commonly used are the L and tee, but a Pi network may be found occasionally. Even a simple shunt fed antenna usually requires a small amount of L or C to complete the match.
Figure 2. A tee network adds an additional series element to an L network.
Figure 1 shows an L network. It consists of a reactor and a capacitor. These networks are simple to design provided that the shunt arm is placed in parallel with the larger resistance or reactance. Apart from design considerations, it doesn't matter which component is placed in the series or shunt position. As might be expected, this network introduces a certain amount of phase shift. Once the network has been built the phase shift and the transformation ratio are fixed. Changing one affects the other, and to overcome this limitation the tee network was developed.
The tee network, shown in Figure 2, has an extra series element. The L network is probably more common in nondirectional AM stations, although I prefer a tee because it allows more latitude when setting up the system. This network is almost always used in the ATUs of directional stations because it provides easier adjustment and phase shift is variable over a range of about 90°.
Tee networks are calculated with specific values for each of the three arms. Often the capacitor value is not a stock item. Therefore, it is common to find a capacitor and inductor in series in any leg. The inductive reactance is used to cancel out the excess capacity and obtain the desired value of C. A vacuum capacitor adjusted to the correct value will sometimes provide better audio than a coil and capacitor combination, but it is more expensive.
The purpose of the phasor is to provide an RF voltage of the correct phase and magnitude to each antenna in an array to produce the desired radiation pattern. Although the base operating impedance of each antenna is important, far more important is the impedance of the Common Point in the phasor. This is the input circuit of the phasor and is the link carrying the combined power of all radiators. The total current squared at that point multiplied by the resistance equals the total power of the transmitter and is known as Icp. This is the point at which the FCC requires power to be measured.
A tee network matches the total phasor load to the transmission line. A typical three-tower system is shown in Figure 3. Immediately following the tee network is a power divider that provides the required power for each tower. These taps provide somewhat coarse magnitude control and the user is advised to specify continuously variable controls. Moving coil taps is time-consuming and requires a lot of patience. The actual type of power distribution circuit usually depends on the preferences of the consulting engineer.
The phase control for each tower drive consists of a tee network in whose series legs are continuously variable inductors coupled to move together. The shunt leg should also use a continuously variable inductor to provide easy adjustment. Normally the reference tower phasor settings are not changed once the pattern has been established correctly.
It sometimes helps to understand the working of the phasor by considering the power branching network as a simple source of RF and seeing individual phasor controls as tee networks matching the transmission lines.
It is important to realize that phasing is the main subject of concern in a directional installation. The consulting engineer calculates the required phase and magnitude of the base current in each tower to produce the desired antenna pattern. Then the phase shift introduced by the transmission line is determined, taking into account the transmission line dielectric factor. A maximum of 90° of phase shift is the usual value for a tee network in the ATU, and the anticipated phase shift in the phasing network is added.
The transmission line length is critical. Any damage to a transmission line or length change will affect the pattern. One of my clients substituted what “he thought was about” the same length of RG-8 to the particular tower and wondered why the pattern was out.
The power divider network is not considered in the phase calculations for the transmission line, phasor and ATUs because it acts only as a power source for the array towers. However, in most power dividers a small change in the power to one antenna will produce a change in the other antennas because the power distribution is altered.
The phasor is not the frightening cabinet of horrors that some engineers consider it to be. If its operation is understood, all dial readings are logged, users remember that a change to one phasor control usually results in at least a minor change in the other towers and the reference tower phasor controls are rarely moved, the phasor can usually be tamed.
E-mail Battison at