Power Amplifiers

1. Class A Power Amplifiers

The purpose of class A bias is to make the amplifier relatively free from distortion by keeping the signal waveform out of the region between 0V and about 0.6V where the transistor’s input characteristic is non linear. Class A design produces good linear amplifiers, but are wasteful of power. The output power they produce is theoretically 50%, but practically only about 25 to 30%, compared with the DC power they consume from the power supply. Class A power amplifiers use the biasing method illustrated in Fig. 5.2.1. This method causes a standing bias current to be flowing during the whole waveform cycle, and even when no signal is being amplified. The standing bias current (the Quiescent Current) is sufficient to make the collector voltage fall to half the supply voltage, and therefore power (P = IC x VCC/2) is being dissipated by the transistor whether any signal is being amplified or not. This was not a great problem in class A voltage amplifiers, where the collector current was very small, but in power amplifiers output currents are thousands of times larger, so efficient use of power is crucial.

Transformer Coupled Class A Output

The circuit shown in fig 5.2.2 is a class A power output stage, but its efficiency is improved by using an output transformer instead of the resistor as its load. The transformer primary winding has high apparent impedance (ZP) at audio frequencies because of the action of the transformer in ‘magnifying’ the impedance of the loudspeaker. As shown by the formula:  

ZP = ZLS (NP/NS)2

The apparent impedance of the primary winding (ZP) will be the actual impedance of the loudspeaker (ZLS) multiplied by the square of the turns ratio. Although the impedance of the transformer primary winding is high, its DC resistance (at 0Hz) is practically zero ohms. Therefore while a class A voltage amplifier might be expected to have a collector voltage of about half supply, a class A power amplifier will have a DC collector voltage approximately equal to the supply voltage (+12V in Fig. 5.2.2) and because of the transformer action, this allows a voltage swing of 12V above and below the DC collector voltage, making a maximum peak to peak signal voltage (Vpp) available of 24V. With no signal, the quiescent collector current of the (medium power) output transistor may typically be about 50mA. When a signal is applied, the collector current will vary substantially above and below this level. Class A power amplifiers, using the relatively linear part of the transistors characteristics are less subject to distortion than other bias classes used in power amplifiers, and although their inferior efficiency improves when output transformers are used, the introduction of a transformer can itself produce additional distortion. This can be minimised by restricting amplitude of the signal so as to utilise less than the full power of the amplifier, but even under optimum conditions the efficiency of class A presents problems. With substantially less than 50% of the power consumed from the supply going into the signal power supplied to the loudspeaker, the wasted power is simply produced as heat, mainly in the output transistor(s). In large high power amplifiers class A is not practical. For example an amplifier used to produce 200W to a large loudspeaker system would need a 400W amplifier producing at its most efficient, 200W of wasted heat that must be dissipated by very large transistors and even larger heat-sinks if overheating, and subsequent component failure is to be avoided. Class A output stages are therefore used mainly in low to medium power output stages of 1 to 2 watt and below, such as domestic radio or TV receivers and headphone amplifiers.

2. Class B Power Amplifiers

In amplifiers using class B bias, illustrated in figure there is no standing bias current (the quiescent current is zero) and therefore the transistor conducts for only half of each cycle of the signal waveform. This dramatically increases efficiency, compared with class A. Theoretically nearly 80% efficiency can be achieved with this bias and in practical circuits, efficiencies of 50% to 60% are possible.   

The downside for this increased efficiency is that the transistor only amplifies half the waveform, therefore producing severe distortion. However, if the other half of the waveform can be obtained in some other way without too much distortion, then class B amplifiers can be used to drive most types of output device. The aim is to obtain a good power gain with as much of the energy consumed from the power supply going into the load as possible. This should be as consistent with reasonable linearity (lack of distortion), as possible. Power output stages do however produce more distortion than do voltage or current amplifiers.

 RF Power Amplifiers Using Class B

The class B bias can be used in a radio frequency (RF) output stage. Although the circuit would produce severe distortion as only half of the signal wave form produces a current in the load, because the load in this case is a tuned circuit resonating at the signal frequency, the resonating effect of the tuned circuit ‘fills in’ the missing half cycles. This method is only suitable at RF, as at lower frequencies the inductors and capacitors needed for the resonant circuit would be too large and costly for most applications. Because of the superior efficiency of class B it is a popular choice for power amplifiers, but to overcome the severe distortion caused by class B, audio amplifiers use a push-pull circuit.

Crossover Distortion

The main problem with class B push pull output stages is that each transistor conducts for NOT QUITE half a cycle. As shown in Fig. 5.3.4 distortion occurs on each cycle of the signal waveform as the input signal waveform passes through zero volts. Because the transistors have no base bias, they do not actually begin to conduct until their base/emitter voltage has risen to about 0.6V. As a result, there is a ‘Dead Zone’ of about 1.2 V around the zero volts line (between -0.6V and +0.6V) where the signal waveform is not amplified, causing a "missing" section from the output signal, resulting in unwanted distortion during the "crossover" from one transistor to the other.

The effect of this distortion on the output depends to some degree on the amplitude of the output signal, the larger the amplitude the less significant the missing 1.2 volts becomes. Also the distortion will be less severe at high frequencies where the rate of change of the wave, as it passes through zero is much faster, causing a shorter ‘step’ in the waveform. The large and varying current drawn by a powerful class B amplifier also puts considerable extra demand on the DC power supply and as the current drawn varies with the amount of signal applied, the smoothing capabilities of the power supply must be efficient enough to prevent this varying current from creating voltage changes at audio frequencies on the power supply lines. If these are not adequately removed, unintended audio feedback into earlier amplifier stages can occur and cause problems with instability. This extra demand on power supply complexity adds to the cost of class B power amplifiers. Crossover distortion is more of a problem in low and medium power class B amplifiers and the method used to eliminate it, is to use a class B amplifier that has some bias (and quiescent current) added so that the output transistors are conducting continually, and so avoiding the ‘dead zone’ of class B. As this method has some properties of both class A and class B it is called Class AB.

3. Class AB Power Amplifiers

The class AB push-pull output circuit is slightly less efficient than class B because it uses a small quiescent current flowing, to bias the transistors just above cut off as shown in figure but the crossover distortion created by the non-linear section of the transistor’s input characteristic curve, near to cut off in class B is overcome. In class AB each of the push-pull transistors is conducting for slightly more than the half cycle of conduction in class B, but much less than the full cycle of conduction of class A. As each cycle of the waveform crosses zero volts, both transistors are conducting momentarily and the bend in the characteristic of each one cancels out. Another advantage of class AB is that, using a complementary matched pair of transistors in emitter Follower mode, also gives cheaper construction. No phase splitter circuit is needed, as the opposite polarity of the NPN and PNP pair means that each transistor will conduct on opposite half cycles of the waveform. The low output impedance provided by the emitter follower connection also eliminates the need for an impedance matching output transformer.

Matching of current gain and temperature characteristics of complementary (NPN/PNP) transistors however, is more difficult than with just the single transistor type as used in class B operation. Also with no emitter resistors, due to the use of emitter follower mode, temperature stability is more difficult to maintain. Class AB therefore, can have a greater tendency towards thermal runaway. The figure illustrates the method of applying the class AB bias to a complementary pair of transistors. The two resistors R1 and R2 apply voltages to the output transistor bases so that Trl (NPN) base is about 0.6V more positive than its emitter, and Tr2 (PNP) base is about 0.6V more negative than its emitter, which is at half of V_CC.

4. Class C Power Amplifiers

In Class C, the bias point is placed well below cut-off as shown in figure and so the transistor is cut-off for most of the cycle of the wave. This gives much improved efficiency to the amplifier, but very heavy distortion of the output signal. Class C is therefore not suitable for audio amplifiers. It is however commonly used in high frequency sine wave oscillators and certain types of RF amplifiers, where the pulses of current produced at the amplifier output can be converted to complete sine waves of a particular frequency by the use of LCR resonant circuits.

5. Class D Power Amplifiers 

In class D audio amplifiers, the basic operation of which is shown in Fig. 5.6.2, the audio signal is first converted to a type of digital signal called ‘Pulse Width Modulation’. This is not a digital signal within the normally accepted definition of ‘Digital’ but only in that it has two levels, high and low. When such a signal is amplified, very little power is dissipated in the amplifier, resulting in much greater efficiency than in conventional analogue amplifiers. The PWM signal is finally converted back into analogue form at the output.

Pulse Width Modulation

The figure illustrates how the audio signal is converted into a ‘pulse width modulated’ form using a comparator, which compares the audio signal, made up of relatively low frequency sine waves, with a much higher frequency triangular waveform. The output of the comparator switches to a high level if the instantaneous voltage of the triangular wave is higher than that of the audio wave, or to a low level if it is lower.The comparator output therefore consists of a series of pulses whose widths vary in relation to the instantaneous voltage of the sine wave. The average level of the PWM signal has the same shape (though inverted in this case) as the original audio signal.