Multi-Stage Amplifier

Lab 6
Salvatore D’Acunto
8508-8811
04/20/2011
 I. OBJECTIVES
The objective of this lab was to design and build a multi-stage amplifier

consisting of a CE amplifier and an EF amplifier biased with a reference current mirror

acting as a current source. The amplifier was used to drive a speaker.

II. THEORY
While single-stage amplifiers are useful on their own, practical amplifiers consist

of multi-stage amplifiers in cascade. The first stage of a multi-stage amplifier is usually

used to provide the majority of the gain as well as provide a high input resistance. A

common emitter amplifier using a BJT is a very good choice for the first stage of this

multi-stage amplifier.

The common emitter amplifier was studied in a previous lab. The equations

governing the current and output characteristics can be seen in equations 1-9 below.

iC = IS*e(VBE/VT) - ISC*e(VBC/VT)

iB = (IS/β)*e(VBE/VT) + ISC*e(VBC/VT)

iC = IS*e(VBE/VT) = α*iE = β*iB

iB = iC/ β =iE/(β + 1)

iE = iC/α = (β + 1)*iB

α = β/(β + 1)

ro = VA / I’C

gm =IC / VT = α/re

Rin = (β +1)*re
 The most common biasing arrangement for the common emitter amplifier is the

classical discrete-circuit bias arrangement. A CE amplifier in this arrangement can be

seen in Figure 1 below.

Figure 1: CE Amplifier in Classical Discrete-Circuit Bias Arrangement

In this biasing arrangement, the emitter current is given by equation 10 below. The

overall voltage gain is given by equation 11.

IE = (VBB –VBE) / (RE + RB /(β +1))

The final stage of a multi-stage amplifier is often used to provide a low output

resistance. This is done to avoid a loss of gain when a low impedance load, such as the

speaker used in this experiment, is connected to the amplifier. The BJT amplifier used to

accomplish this is the emitter follower, or common collector amplifier. The EF amplifier

acts as a voltage buffer. The voltage gain of an ideal EF amplifier is unity. The output

resistance is given by equation 12 below.

RO = re
 Biasing a BJT using a constant current source is very effective. This is because

the current source allows the emitter current to be independent of the values of β and RB.

A current source can be created using two BJTs. This configuration is known as a

current mirror. This configuration can be seen in Figure 2 below.

Figure 2: Current Mirror

In the above figure, IREF is the current given by equation 13 below flowing into the

collector of Q1.

IREF = (VCC –VBE) / RREF

When the two BJTs are on the same IC, Q1 and Q2 are generally the same. This

allows for their collector currents to be equal, hence the term current mirror. This means

that IOUT = IREF.

 III. Results
In this experiment, a CE amplifier was cascaded with an EF amplifier biased with

a current mirror. The multi-stage amplifier can be seen in Figure 3 below.

Figure 3: Multistage Amplifier

The amplifier was designed to meet the following specifications.

1. 100mV Peak at 10kHz

2. RSIG = 100Ω

3. RL = 8Ω

4. Minimal Distortion at Output

5. All Resistors Standard Values

6. No Resistor Greater Than 100kΩ

7. All Capacitors 2.2μF

The component theoretical and measured values can be seen in Table 1 below.

RSIG 100Ω 100.1Ω

RB1 100kΩ 100.7kΩ

RB2 50kΩ 49.3kΩ

RC 15kΩ 14.7kΩ

RE1 220Ω 219.3Ω

RE2 8.3kΩ 8.3kΩ

RL 10Ω 10.1Ω

ZL 8Ω 8Ω
RREF 2kΩ 2.1kΩ

RE3 220Ω 221.2Ω

C1 2.2μF 2.2μF

C2 2.2μF 2.2μF

C3 2.2μF 2.2μF

Table 1: Component Values

In the first part of the experiment, the load was a 10Ω resistor. In the second part

of the experiment the 10Ω resistor was replaced with a speaker. The speaker had an

impedance of approximately 8Ω, but is a series impedance of a resistor and an inductor.

This potentially leads to distortion in the output waveforms. The voltage waveforms for

the first and second parts of the experiment can be seen in Figures 4 and 5 below

respectively. The gain in each case was approximately .62 V/V and .67 V/V,
Figure 4: Voltage Waveform for 10kΩ Resistor Load

Figure 5: Voltage Waveform for 8Ω Speaker Load

 The DC operating points for all of the transistor nodes were calculated using the

equations in the theory section. The results can be seen in Table 2 below. The subscript

number corresponds to the number of the transistor in the figure above.

VC1 7.0V IC1 0.53mA

VC2 15V IC2 7.0mA
VC3 4.7V IC3 7.07mA
VC4 0.7V IC4 7.07mA
VB1 4.9V IB1 0.01mA
VB2 7.0V IB2 0.07mA
VB3 0.7V IB3 0.07mA
VB4 0.7V IB4 0.07mA
VE1 4.2V IE1 0.54mA
VE2 6.3V IE2 7.07mA
VE3 0V IE3 7.14mA
VE4 0V IE4 7.14mA
Table 2: DC Operating Points for Transistors

In a multi-stage amplifier, the input resistance is the input resistance from the first

stage. The output resistance likewise, is the output resistance of the final stage. The

overall voltage gain is the product of the voltage gain from each stage because the

amplifiers are cascaded together. These values are seen below in Table 3.

ZIN ZOUT GV
518.3Ω 3.5Ω 4.82 V/V
Table 3: Characteristics

IV. DISCUSSION

3.2 CE Amplifier

The CE Amplifier was the first stage in a multi-stage amplifier used to drive a

load. The typical load in audio applications is a speaker. The theoretical value for the

voltage gain and the experimental value were very different. The percent error was

approximately 86.1%. There are several reasons that this error is expected. When
calculating the theoretical values, the value of β was assumed to be 100 for each

transistor. This is a value that has a high variation among transistors. This is the reason

biasing using schemes were the amplifiers are not dependant on β is important. This type

of biasing was used in the EF stage but not in the CE stage where most of the gain is

achieved. Also there was distortion in both graphs. There may have been a problem with

one or more of the BJTs since the same distortion occurred for the resistor and speaker.

A speaker has impedance that consists of both resistance and inductance. The overall

voltage gain of a CE amplifier is directly proportional to the parallel resistance of the

collector resistor and the load resistor. At open circuit, the load resistance is infinite and

the gain is directly proportional to the collector resistor. When the load becomes small,

as in the case of an 8Ω speaker, the parallel resistance will always be smaller than both of

the two resistors and thus the gain will also decrease. One way to minimize the

difference in overall voltage gain is to reduce the value of the collector resistor. By

choosing a very low value, you will ensure that the parallel resistance is always a close

value to RC. Even if zero sensitivity were realizable, this would mean that RC would be

very small and close to zero. Since the voltage gain is proportional to RC, the gain would

also be very small or close to zero. This is the reason for multi-stage amplifiers to handle

small loads.

4.1 Two-Stage Amplifier

Multi-stage amplifiers are very beneficial for providing important features of

several types of amplifiers. However, even these types of amplifiers have limitations.

The Q2 and Q4 transistors restricted the output voltage swing of the amplifier. At the
emitter side of Q2, the voltage at the emitter limits the maximum voltage swing at the

output where Q2 will enter the cutoff region of operation. At the collector end of Q4, the

maximum negative voltage swing allowable is the lowest voltage of the collector before

Q4 leaves the active region and enters the saturation region of operation. The EF

amplifier is best used in the circuit as the second stage of the amplifier when connecting

the 8Ω load. This is because the EF amplifier acts as a voltage buffer. It provides a very

high input resistance at the output of the CE amplifier and thus reduces the loss in gain

from the first stage. However, it also has a very low output resistance at the load. This

allows the lowest loss of gain at the load while providing unity gain from the voltage

from the first stage. Although useful, the EF amplifier is a Class A amplifier stage and

thus has high power dissipation even when there is no input or output. To overcome this

problem power BJTs can be used, as in an LM386. The LM386 has three stages of

amplifiers. The first stage is a common emitter amplifier consisting of Q1 and Q2. The

second stage is the differential amplifier consisting of Q3 and Q4. The final stage is the

Class AB power amplifier consisting of Q7, Q8, and Q9.

V. Summary
In conclusion, electronic circuit components are necessary to the technology of

today’s world. Linear circuit components are an indispensable class of components but

much of the circuitry in this new age requires nonlinear devices, such as diodes, BJTs,

and MOSFETs. These linear and nonlinear devices can be integrated together to form

amplifier circuits that have many practical implications. To further increase the usability

of these amplifiers, multi-stage amplifiers can be developed from cascading single stage
amplifiers. As device technology continues to grow, so will the complexity of our

technology driven world.