Project 19
Push-Pull (Class AB, B) Amplifiers

Objective: This project will investigate the operation of a push-pull amplifier. Class B and Class AB operation will be examined.

Components: 2N2222 BJT, 2N2907

Introduction:

Many small signal amplifiers, both BJT and FET, are biased to provide symmetric output voltage swings. This bias arrangement, as examined in many of the previous projects, requires a DC collector (drain) current and a DC collector-emitter (drain-source) voltage even when no input signal is being amplified. The product of this bias current and voltage results in the transistor dissipating power at all times. The class of amplifiers that have a continuous power dissipation are referred to as Class A amplifiers. The Push -Pull amplifier illustrated in Figure 19-1 is referred to as a Class B amplifier since the transistors are off unless an input signal is present and there is one transistor used for positive and one transistor used for negative output voltage swings. In this arrangement, there is no DC power dissipated by the transistor. The transistors illustrated in Figure 19-2 have a very small bias applied to the each base to improve the linearity of the amplifier and reduce crossover distortion. This type of arrangement is referred to as Class AB since it has characteristics of both Class A and Class B amplifiers. There are many additional amplifier classes including C, D, E, F, G, H, and S, each of which is designed for specific operating conditions. The majority of these additional amplifier types rely on frequency dependent input and/or output matching networks for proper operation and are beyond the scope of this project. The operation of the Class B and Class AB amplifier will be the focus of this project.

The Push-Pull amplifier is designed to provide large output voltage swings and power gain. The two transistors are arranged to allow one (NPN) to provide the current drive for positive swings and the other (PNP) to provide the drive for the negative swings. The NPN transistor is said toSource current while the PNP is said to Sink current. This Source-Sink relationship leads to the designation as a Push (source) - Pull (sink) amplifier. The circuit in Figure 19-1 requires a non-zero input signal ( VBE(on)) to be present before either transistor turns on. This results in a distorted output signal since there is no output signal when the input signal is between  - 700 mV and  + 700 mV. The non-linearity (distortion) in the output signal is referred to as Crossover Distortion since it occurs when the input signal crosses over from positive to negative or negative to positive. The amplifier shown in Figure 19-2 is designed to prevent the crossover distortion by maintaining the two transistors at the Edge of Conduction (EOC). Ideally, this arrangement allows each transistor to respond to any non-zero input. The actual design however usually aims to maintain the two transistors in a slightly forward biased operating state. Since each transistor has a small DC bias it will dissipate a small DC power even in the absence of an input signal. The Class AB designation is a consequence of this power dissipation. Care must be taken in the Class AB design not to have a large "reverse" crossover distortion by having both transistors operating for a given input signal.

Figure 19 - 1: Class B Push-Pull Amplifier

Figure 19 - 2: Class AB Push-Pull Amplifier

Design:

1. Design a Class B amplifier as illustrated in Figure 19-1 that can drive a 2 k load with
a ± 10 V output swing. Determine the gain for the amplifier. You may assume you have between ± 15 V and ± 20 V power supplies available. Verify your design using PSPICE®.

2. Repeat step 1 for the Class AB amplifier illustrated in Figure 19-2. You may use 400  as a starting point for the value of R. Be careful in your PSPICE® analysis to use a diode that can handle the DC current.

Lab Procedure:

1. Construct the amplifier designed in step 1 of the design procedure. Verify proper operation of the design.

2. Determine the amplifier voltage, current, and power gains along with the maximum symmetric (effectively) output swing possible for your design.

3. Adjust the input signal to produce a 5 V peak output wave and measure the crossover distortion in terms of input voltage levels and the phase angle(s) over which the output wave is zero. Repeat for a 2 V peak output wave.

4. Observe the loading affect by replacing RL first by 150  and then by 15 k. Note any changes in the output signal and comment on the loading affect.

5. Use computer control to record and plot the frequency response. Find the corner frequencies and bandwidth.

6. Measure the input impedance seen by the source [look at the current from the function generator and the node voltage at the base of the transistors] and the output impedance seen by the load resistor [look at the open circuit voltage and the current through and voltage across RL = 2 k].

7. Repeat steps 1 - 6 for the Class AB amplifier designed in step 2 of the design procedure.

Questions:

1. Compare the output waveforms for the two circuits and discuss the similarities and differences especially in terms of crossover distortion.

2. The two designs do not have input or output coupling capacitors. Why are they omitted from these designs? Would the operation be improved if they were included?

3. Could either (or both) of these amplifiers be designed using a single power supply? Justify your answer.

Do the two transistors have to be "matched", i.e. equal device parameters such as  and VBE(on), to operate properly. Explain your answer.