What is Amplitude Modulators?
Amplitude modulators are generally one of two types: low level or high level. Low-level modulators generate AM with small signals and thus must be amplified considerably if they are to be transmitted. High-level modulators produce AM at high power levels, usually in the final amplifier stage of a transmitter. Although the discrete component circuits discussed in the following sections are still used to a limited extent, keep in mind that today most amplitude modulators and demodulators are in integrated-circuit form.
Low Level Amplitude Modulators
One of the simplest amplitude modulators is the diode modulator described in Sec. 4-1. The practical implementation shown in Fig. 4-7 consists of a restive mixing network, a diode rectifier, and an LC tuned circuit. The carrier (Fig. 4-8b) is applied to one input resistor and the modulating signal (Fig. 4-8a) to the other. The mixed signals appear across R3. This network causes the two signals to be linearly mixed, i.e., algebraically added. If both the carrier and the modulating signal are sine waves, the waveform resulting at the junction of the two resistors will be like that shown in Fig. 4-8(c), where the carrier wave is riding on the modulating signal. This signal is not AM. Modulation is a multiplication process, not an addition process. The composite waveform is applied to a diode rectifier. The diode is connected so that it is forward-biased by the positive-going half-cycles of the input wave. During the negative portions of the wave, the diode is cut off and no signal passes. The current through the diode is a series of positive-going pulses whose amplitude varies in proportion to the amplitude of the modulating signal [see Fig. 4-8(d)].
These positive-going pulses are applied to the parallel-tuned circuit made up of L and C, which are resonant at the carrier frequency. Each time the diode conducts, a pulse of current flows through the tuned circuit. The coil and capacitor repeatedly exchange energy, causing an oscillation, or “ringing,” at the resonant frequency. The oscillation of the tuned circuit creates one negative half-cycle for every positive input pulse. Highamplitude positive pulses cause the tuned circuit to produce high- amplitude negative pulses. Low-amplitude positive pulses produce corresponding low-amplitude negative pulses. The resulting waveform across the tuned circuit is an AM signal, as Fig. 4-8(e) illustrates. The Q of the tuned circuit should be high enough to eliminate the harmonics and produce a clean sine wave and to filter out the modulating signal, and low enough that its bandwidth accommodates the sidebands generated.
This signal produces high-quality AM, but the amplitudes of the signals are critical to proper operation. Because the nonlinear portion of the diode’s characteristic curve occurs only at low voltage levels, signal levels must be low, less than a volt, to produce AM. At higher voltages, the diode current response is nearly linear. The circuit works best with millivolt-level signals.
An improved version of the circuit just described is shown in Fig. 4-9. Because it uses a transistor instead of the diode, the circuit has gain. The emitter-base junction is a diode and a nonlinear device. Modulation occurs as described previously, except that the base current controls a larger collector current, and therefore the circuit amplifies. Rectification occurs because of the emitter-base junction. This causes larger half-sine pulses of current in the tuned circuit. The tuned circuit oscillates (rings) to generate the missing half-cycle. The output is a classic AM wave.
A differential amplifier modulators makes an excellent amplitude modulator. A typical circuit is shown in Fig. 4-10(a). Transistors Q1 and Q2 form the differential pair, and Q3 is a constant-current source. Transistor Q3 supplies a fixed emitter current IE to Q1 and Q2, one-half of which flows in each transistor. The output is developed across the collector resistors R1 and R2. The output is a function of the difference between inputs V1 and V2; that is, Vout = A(V2 – V1), where A is the circuit gain. The amplifier can also be operated with a single input. When this is done, the other input is grounded or set to zero. In Fig. 4-10(a), if V1 is zero, the output is Vout A(V2). If V2 is zero, the output is Vout= A(-V1) = -AV1. This means that the circuit inverts V1. The output voltage can be taken between the two collectors, producing a balanced, or differential, output. The output can also be taken from the output of either collector to ground, producing a single-ended output. The two outputs are 180° out of phase with each other. If the balanced output is used, the output voltage across the load is twice the single-ended output voltage. No special biasing circuits are needed, since the correct value of collector current is supplied directly by the constant-current source Q3 in Fig. 4-10(a). Resistors R3, R4, and R5, along with VEE, bias the constant-current source Q3. With no inputs applied, the current in Q1 equals the current in Q2, which is IE/2. The balanced output at this time is zero. The circuit formed by R1 and Q1 and R2 and Q2 is a bridge circuit. When no inputs are applied, R1 equals R2, and Q1 and Q2 conduct equally. Therefore, the bridge is balanced and the output between the collectors is zero.
Now, if an input signal V1 is applied to Q1, the conduction of Q1 and Q2 is affected. Increasing the voltage at the base of Q1 increases the collector current in Q1 and decreases the collector current in Q2 by an equal amount, so that the two currents sum to IE. Decreasing the input voltage on the base of Q1 decreases the collector current in Q1 but increases it in Q2. The sum of the emitter currents is always equal to the current supplied by Q3. The gain of a differential amplifier is a function of the emitter current and the value of the collector resistors. An approximation of the gain is given by the expression A = RC IE /50. This is the single-ended gain, where the output is taken from one of the collectors with respect to ground. If the output is taken between the collectors, the gain is two times the above value. Resistor RC is the collector resistor value in ohms, and IE is the emitter current in milliamperes. If RC = R1 = R2 = 4.7 kV and IE = 1.5 mA, the gain will be about A = 4700(1.5)/50 = 7050/50 = 141. In most differential amplifiers, both RC and IE are fixed, providing a constant gain. But as the formula above shows, the gain is directly proportional to the emitter current. Thus if the emitter current can be varied in accordance with the modulating signal, the circuit will produce AM. This is easily done by changing the circuit only slightly, as in Fig. 4-10(b). The carrier is applied to the base of Q1, and the base of Q2 is grounded. The output, taken from the collector of Q2, is single-ended. Since the output from Q1 is not used, its collector resistor can be omitted with no effect on the circuit. The modulating signal is applied to the base of the constant-current source Q3. As the intelligence signal varies, it varies the emitter current. This changes the gain of the circuit, amplifying the carrier by an amount determined by the modulating signal amplitude. The result is AM in the output. This circuit, like the basic diode modulator, has the modulating signal in the output in addition to the carrier and sidebands. The modulating signal can be removed by using a simple high-pass filter on the output, since the carrier and sideband frequencies are usually much higher than that of the modulating signal. A bandpass filter centered on the carrier with sufficient bandwidth to pass the sidebands can also be used. A parallel tuned circuit in the collector of Q2 replacing RC can be used. The differential amplifier makes an excellent amplitude modulators. It has a high gain and good linearity, and it can be modulated 100 percent. And if high-frequency transistors or a high-frequency IC differential amplifier is used, this circuit can be used to produce low-level modulation at frequencies well into the hundreds of megahertz. MOSFETs may be used in place of the bipolar transistors to produce a similar result in ICs.
Amplifying Low-Level AM Signals.
In low-level Amplitude Modulators circuits such as those discussed above, the signals are generated at very low voltage and power amplitudes. The voltage is typically less than 1 V, and the power is in milliwatts. In systems using low-level modulation, the AM signal is applied to one or more linear amplifiers, as shown in Fig. 4-11, to increase its power level without distorting the signal. These amplifier circuits—class A, class AB, or class B—raise the level of the signal to the desired power level before the AM signal is fed to the antenna.
High Level Amplitude Modulators
In high-level AM, the modulators varies the voltage and power in the final RF amplifier stage of the transmitter. The result is high efficiency in the RF amplifier and overall high-quality performance.
One example of a high-level Amplitude Modulators circuit is the collector modulator shown in Fig. 4-12. The output stage of the transmitter is a high-power class C amplifier. Class C amplifiers conduct for only a portion of the positive half-cycle of their input signal. The collector current pulses cause the tuned circuit to oscillate (ring) at the desired output frequency. The tuned circuit, therefore, reproduces the negative portion of the carrier signal (see article 7 for more details).
The modulator is a linear power amplifier that takes the low-level modulating signal and amplifies it to a high-power level. The modulating output signal is coupled through modulation transformer T1 to the class C amplifier. The secondary winding of the modulation transformer is connected in series with the collector supply voltage VCC of the class C amplifier. With a zero-modulation input signal, there is zero-modulation voltage across the secondary of T1, the collector supply voltage is applied directly to the class C amplifier, and the output carrier is a steady sine wave. When the modulating signal occurs, the ac voltage of the modulating signal across the secondary of the modulation transformer is added to and subtracted from the dc collector supply voltage. This varying supply voltage is then applied to the class C amplifier, causing the amplitude of the current pulses through transistor Q1 to vary. As a result, the amplitude of the carrier sine wave varies in accordance with the modulated signal. When the modulation signal goes positive, it adds to the collector supply voltage, thereby increasing its value and causing higher current pulses and a higher-amplitude carrier. When the modulating signal goes negative, it subtracts from the collector supply voltage, decreasing it. For that reason, the class C amplifier current pulses are smaller, resulting in a lower-amplitude carrier output. For 100 percent modulation, the peak of the modulating signal across the secondary of T1 must be equal to the supply voltage. When the positive peak occurs, the voltage applied to the collector is twice the collector supply voltage. When the modulating signal goes negative, it subtracts from the collector supply voltage. When the negative peak is equal to the supply voltage, the effective voltage applied to the collector of Q1 is zero, producing zero carrier output. This is illustrated in Fig. 4-13. In practice, 100 percent modulation cannot be achieved with the high-level collector modulator circuit shown in Fig. 4-12 because of the transistor’s nonlinear response to small signals. To overcome this problem, the amplifier driving the final class C amplifier is collector-modulated simultaneously. High-level modulation produces the best type of Amplitude Modulation, but it requires an extremely high-power modulator circuit. In fact, for 100 percent modulation, the power supplied by the modulator must be equal to one-half the total class C amplifier input power. If the class C amplifier has an input power of 1000 W, the modulator must be able to deliver one-half this amount, or 500 W.
Example 4-1 An Amplitude Modulators transmitter uses high-level modulation of the final RF power amplifier, which has a dc supply voltage VCC of 48 V with a total current I of 3.5 A. The efficiency is 70 percent.
A major disadvantage of collector modulators is the need for a modulation transformer that connects the audio amplifier to the class C amplifier in the transmitter. The higher the power, the larger and more expensive the transformer. For very high power applications, the transformer is eliminated and the modulation is accomplished at a lower level with one of the many modulator circuits described in previous sections. The resulting AM signal is amplified by a high-power linear amplifier. This arrangement is not preferred because linear RF amplifiers are less efficient than class C amplifiers. One approach is to use a transistorized version of a collector modulator in which a transistor is used to replace the transformer, as in Fig. 4-14. This series modulator replaces the transformer with an emitter follower. The modulating signal is applied to the emitter follower Q2, which is an audio power amplifier. Note that the emitter follower appears in series with the collector supply voltage +VCC. This causes the amplified audio modulating signal to vary the collector supply voltage to the class C amplifier Q1, as illustrated in Fig. 4-13. And Q2 simply varies the supply voltage to Q1. If the modulating signal goes positive, the supply voltage to Q1 increases; thus, the carrier amplitude increases in proportion to the modulating signal. If the modulating signal goes negative, the supply voltage to Q1 decreases, thereby decreasing the carrier amplitude in proportion to the modulating signal. For 100 percent modulation, the emitter follower can reduce the supply voltage to zero on maximum negative peaks.
Using this high-level modulating scheme eliminates the need for a large, heavy, and expensive transformer, and considerably improves frequency response. However, it is very inefficient. The emitter-follower modulator must dissipate as much power as the class C RF amplifier. For example, assume a collector supply voltage of 24 V and a collector current of 0.5 A. With no modulating signal applied, the percentage of modulation is 0. The emitter follower is biased so that the base and the emitter are at a dc voltage of about one-half the supply voltage, or in this example 12 V. The collector supply voltage on the class C amplifier is 12 V, and the input power is therefore
Pin = VCCIc = 12(0.5) = 6 W
To produce 100 percent modulation, the collector voltage on Q1 must double, as must the collector current. This occurs on positive peaks of the audio input, as described above. At this time most of the audio signal appears at the emitter of Q1; very little of the signal appears between the emitter and collector of Q2, and so at 100 percent modulation, Q2 dissipates very little power. When the audio input is at its negative peak, the voltage at the emitter of Q2 is reduced to 12 V. This means that the rest of the supply voltage, or another 12 V, appears between the emitter and collector of Q2. Since Q2 must also be able to dissipate 6 W, it has to be a very large power transistor. The efficiency drops to less than 50 percent. With a modulation transformer, the efficiency is much greater, in some cases as high as 80 percent. This arrangement is not practical for very high power AM, but it does make an effective higher-level modulator for power levels below about 100 W.