Lab 4 - Operational Amplifiers
Lab 4 - Operational Amplifiers
EE052
PRINCIPLES OF EE1 LAB
Lab 4
Operational Amplifiers
I. OBJECTIVES
1. To introduce operational amplifiers and dependent sources
2. To explore those circuit connections that allow operational amplifiers to operate in
their linear region.
II. INTRODUCTION
Ideal operational amplifiers (Op-Amps) are two-ports that can produce an output
voltage which is directly proportional to their input voltage (linear operation). Op-Amps can
be operated in two ways: open loop and closed loop. The latter circuit connection is the only
one that can force the Op-Amp to operate in its linear region. An equivalent circuit model can
be used to model or simulate the ideal Op-Amp or to incorporate deviations from ideality. The
standard inverting and non-inverting configurations are explored.
The lab experiments include the realization of both configurations and the
experimental determination of the circuit parameters that demonstrate the function of the
circuit and allow for Op-Amp parameter derivation.
1. Dependent Sources
Dependent sources are sources whose value varies as a function of a specified voltage
or current elsewhere in the circuit. The relationship could be of any form, but in this course
we will introduce only those sources whose value is proportional to a voltage or current
elsewhere in the circuit. Since the output quantity can be voltage or current and so can the
controlling quantity, there are four types of such dependent sources, whose names,
characteristic equations, and symbols are shown in Fig. 1.
2. Operational Amplifiers
A 741 Op-Amp is shown in Fig. 2 below. Op-Amps have two input terminals; the
input voltage Vi to the Op-Amps is taken across these terminals. One terminal is called
inverting or negative and the voltage there is usually denoted as V n and the other as
noninverting (Vp) so that Vi=(Vp-Vn). The output is taken between Vo and ground. Additional
terminals (such as +Vcc, or -Vcc) are used for bias, offset etc.
The realistic model of an operational amplifier is given in your text and repeated
below with equivalent notation. It involves separate input and output circuits. The input
consists of an input resistance R i between the inverting and noninverting terminals. The
output consists of a voltage dependent voltage source (with voltage A vVi) in series with an
output resistance Ro. Note that the only connection between the input and output is through
the proportionality relation of the dependent source.
ii. Output Voltage Vo: The output voltage of an Op-Amp is proportional to the input
voltage, provided it remains less in absolute value than the DC bias voltages Vcc
and -Vcc.
iii. Input Resistance Ri : The input resistance appears between the inverting and
noninverting terminal (so that Vi appears across Ri) and can be found by dividing
the input voltage Vi by the current entering the non-inverting input terminal V p or
exiting the inverting terminal Vn.
iv. Open Loop Voltage Gain μ or Av: The open loop voltage gain is the
proportionality constant in the dependent source equation where V = A vVi (or
V=μV(a,b)).
v. Output Resistance Ro: The output resistance appears as a resistor in series with
the dependent source. In the presence of a non-zero output resistance R o, the
output voltage across a load RL is not all of V = A vVi and can be found by
analyzing the voltage divider between Ro and RL.
Op-Amps have two regions of operation: linear and saturation. In the linear region,
the voltage transfer characteristic, i.e. the mathematical relationship between the input and
output voltages, is linear. This holds true when the output voltage lies in the range
-Vcc ≤ V0 ≤ Vcc
From the definition of voltage gain given above, i.e. Vo = AvVi, one can see that this
range corresponds to input voltages in the range of
−V cc + V cc
≤ V i≤
Av Av
In this range the output voltage is directly proportional to the input voltage, by the
factor Av.
For input voltages outside this range, the Op Amp is said to be saturated, and its
output is bounded by the DC bias voltages. In other words, the output voltage is clamped to
-Vcc when Vi < -Vcc/Av and to Vcc when Vi > Vcc/Av.
ii. Ro = 0: In this case the entire dependent source voltage appears across the load
resistance or as the input of another device.
iii. μ =AV = : If the output voltage is to be finite it follows from the definition of
voltage gain, that Vi = Vo / Av will go to zero if Av is infinite. This, however,
assumes that there is some way for the input to be affected by the output. Indeed
this will only happen if there is such a connection namely a negative feedback
mechanism in the form of a connection between the output and the inverting
terminal (closed loop operation). If such connection does not exist, then the output
will be saturated (open loop operation). For closed loop operation, it is said that a
virtual short exists between the positive and negative input terminals. This means
that if an Op-Amp is operating in its linear region (if it is unsaturated) then Vi 0,
or equivalently Vp Vn. This also simplifies the circuit calculations at the input
terminals, because Vp and Vn can be represented by a single variable. When one of
the two terminals is grounded, then the voltage at both terminals is zero and the
other terminal is called a virtual ground.
There are two standard closed-loop connections for an Op-Amp. Both have in
common the connection (Rf) from the output terminal to the inverting input terminal. This
connection provides the negative feedback and ensures the virtual short. The analysis is
simple for ideal Op-Amps since:
The analysis usually derives a gain or amplification. It is important to note that this is
the gain of the whole stage (or the closed loop gain) and should not be confused with the gain
of the Op Amp alone.
One last note: negative feedback does not guarantee that the amplifier will not
saturate. If the input is such that the output, based on the amplification of the whole stage, is
expected to be larger than the bias voltage in absolute value (Vo> +Vcc or Vo< -Vcc) then the
output will be clamped to Vcc (or -Vcc).
Thus, the theoretical gain K of the whole stage (that is, the entire Op-Amp circuit of
Fig 4.) is given by
Circuit analysis of the non-inverting amplifier shown in Fig. 5 yields the equation,
K = V2/V1= (1 + Rf / R).
Figure 6
Using a VCVS, one can construct a model of the Op-Amp for use in SPICE. The
circuit of Fig. 2b can be used to model a non-ideal1 Op-Amp using two resistors and a
dependent voltage source.
The circuit of Fig. 6 can be used for simulating an ideal Op Amp and is derived from
Fig. 2b by shorting out the output resistor R o (which is equivalent to setting its value equal to
zero) and by picking large values for the input resistor R i and for the Op-Amp voltage gain μ
(or A). Typical such values for approximating an ideal Op-Amp in PSpice are Ri=10 10 and μ
=106.
III. PRE-LABRATORY
1. Theory
i. Briefly explain why one can assume Vp=Vn for an ideal Op-Amp. What
connection has to be present for this to occur?
ii. What is the gain of an amplifier circuit? How is it different from the Op-Amp
gain?
2. Experiment 1
i. Calculate the gain K for the non-inverting amplifier circuit in Fig. 8 (from
section 5.1 below) assuming that the Op-Amp is ideal and using the resistance
values specified in 5.1.1.
ii. Calculate the theoretical range of the input voltage for linear operation of the
circuit in Section 5.1.
iv. The PSpice Op Amp model presented in Section 3.2.5 does not account for the
effects of saturation, so this portion of the experiment cannot be simulated in
PSpice. Describe how you would expect the circuit to behave outside its range
of linear operation.
3. Experiment 2
i. Calculate the gain K for the inverting amplifier circuit of Fig. 9 (from Section
5.2 below) assuming that the Op-Amp is ideal. The answer should be in terms
of R and Rf.
ii. Given the results of question 4.7, calculate the values of R and R f that produce
a circuit gain of -4.545 and a voltage Vi=0.5V when Vs=5V.
iii. Simulate the experimental procedure from Section 5.2 in PSpice by choosing
3 different points in the linear operating range, and calculating the circuit gain
at each of these points.
Digital Multimeter
V. PROCEDURES
You will be using the "741" Op-Amp which is biased at +15V and -15V. The chip
layout is shown in Fig. 7. The standard procedure on such chip packages (DIP15) is to
identify pin 1 as the one to the left of the notch in the chip package. The notch always
separates pin 1 from the last pin on the chip. In the case of 741, the notch is between pins 1
and 8. Pins 2, 3, and 6 are the inverting input V n , the non-inverting input Vp, and the
amplifier output Vo respectively. These three pins are the only three terminals that usually
appear in an Op-Amp circuit schematic diagram.
Figure 8
2. Use the fixed 5V power supply of the power source for V s. Vary Rvar’s value so that
you can change Vi. Take readings for the output voltage V out for values of Vi from -
3.5V to +3.5V in increments of 0.5V and record them in Table 1. Calculate KV i for
each Vi using the calculated gain K found in prelab item 4.3 above. Calculate the %
error for each row in the table.
0 0 0 0 0%
3. For an input voltage of your choice that keeps the Op-Amp in the linear region,
place an ammeter in series with Rf. Record the value of the current I.
Vi = ____________ . I = ______________ .
4. Disconnect the ammeter. Keep the input voltage the same as in 5.1.3 above. Place a
10k load resistor between the output terminal of the Op-Amp and ground. In so
doing one can study the output resistance characteristics of the Op-Amp. Measure
the output voltage Vout with the DVM, and compare with the results obtained for
the same input voltage in item 5.1.2. Explain any discrepancies by assuming a non-
zero Op-Amp output resistance. Later you will be asked to calculate the output
resistance of the Op Amp based on these results.
5. This item involves the study of the relationship between the load resistance and
output voltage (and thus also voltage gain). Keeping the source voltage at 5V,
measure IL (the current through the load resistance R L) for each value of RL in Table
2. Later, you will be asked to analyze this data.
10k 0.145mA
15k 0.1022m
20k 0.0547m
Figure 9
1. In prelab item 4.7 you should have calculated the values of Rf and R that yield a circuit
gain of -4.545 and Vi=.5V when Vs=5V and Rvar=20k. Get your TA to check your
calculations and correct them if necessary, then build the circuit of Fig. 9 with the
correct values of Rf and R.
2. Use the fixed 5V power supply of the power source for V s. Vary Rvar’s value so that
you can change Vi. Take 21 readings for the output voltage V out at each value of Vi
from -5V in increments of 0.5V and record them in Table 3. Calculate KVi for each V i
in Table 3 using the calculated gain K found in prelab item 4.7 above. Calculate the %
error for each row in the table. If the measured Vout differs from KVi by more than 10%
you probably have an error in the circuit. Troubleshoot the circuit until it is operating
properly.
0 0 0 0 0%
VI. REPORT
1. Derive the relationship between the current I and the resistor R f in the non-inverter
circuit of Fig. 8.
2. Compare the theoretical value of the gain K = Vout /Vi of both the inverting and the
non-inverting circuits of Sections 5.1 and 5.2 that you calculated in the prelab
exercises with the experimentally obtained values of gain.
3. Calculate the theoretical value of the current I for the resistor R f in Section 5.1.
Compare with the experimental one.
4. Calculate the theoretical values of the current I L for all three values of RL in Section
5.1.5. Compare with the experimental ones.
5. Plot the experimental values of IL vs. 1/RL in a graph with rectangular coordinates.
From your graph, how does your output voltage depend on the load? How does the
gain K= Vout /Vi depend on the load? Note that if V out does not change with the load
RL, and since IL = Vout (1/RL), then the slope is Vout and it should be constant and thus
the graph of IL vs. 1/RL should be a straight line passing through the origin.
6. Draw two graphs of the experimentally obtained V i vs. Vout, one for the inverting
amplifier circuit and one for the non-inverting amplifier circuit (5.1 and 5.2). On each
graph identify the transition between saturated and linear regions of operation for
these amplifier circuits. Label the mode of operation for each of these regions. For the
linear regions and for both circuits, discuss the possible sources of discrepancies
between the experimentally obtained value of Vout and the calculated values of KVi.
7. Simulate the non-inverter circuit of Fig. 8 in PSpice for Rf = 10k, Rvar=20 k and
R=2.2 k. Find the output voltage Vout and the current I in Rf. Assume a μA741 Op
Amp.
8. Simulate the non-inverter circuit of Fig. 8 in PSpice for Rf = 10k with a load RL =
10k applied between the output terminal of the Op-Amp and ground. Find the
current in RL.
9. Simulate the inverter circuit in Fig. 9 in PSpice for Rf = 10k, Rvar=20k and R=2.2
k. Find the output voltage Vout and the current in Rf.