The full analysis of the op-amp circuits as shown in the three examples
in the previous section may not be necessary if only the voltage gain is
of interest, which can be found based on the “virtual ground” assumption.
Specifically, as
is in the range between the positive
and negative voltage supplies (e.g.,
, the rails) and
,
, i.e.,
.
If one of the two inputs is grounded, the other one is also approximately
grounded, or virtually grounded. More generally,
and
can
be assumed to be virtually the same, even if none of them is grounded. Based
on this assumption, the analysis of all op-amp circuits is significantly
simplified. However, note that if the input and output resistances of the
op-amp circuits are of interest as well, the full analysis shown previously
is necessary.
Now consider the voltage gain of the following typical op-amp circuits:
(34) |
As is very large, the current into the op-amp is negligible, and
. Applying KCL to the node of
, we have
(35) |
In general, and
of the inverter can be replaced by two networks (with
impedances
and
respectively) containing resistors and capacitors and
the analysis of the circuit can be carried out easily in frequency domain:
(36) |
Also, the input resistance can be easily obtained to be
, based on
the virtual ground assumption.
(37) |
Apply KCL to :
(38) |
We note that the differential amplifier is similar to an inverting
amplifier but with an additional input to the non-inverting side.
We first define
, and then apply KCL to both
and
to get:
(39) |
(40) |
(41) |
Typically, we let and
, and get
(42) |
(43) |
(44) |
The input resistances of this differential amplifier can be found to be
(45) |
(46) |
Note that there is a conflict of interest in terms of . To have
a large gain
, we want
to be small, but to have a high
input resistance
, we want
to be large. This issue will be
addressed later when we discuss the instrumentation amplifier.
We further consider some special cases:
(47) |
(48) |
(49) |
(50) |
If one of the two inputs, connected to a constant reference voltage
treated as a reference voltage, then the differential amplifier
can also be used as a level shifter.
(51) |
(52) |
It can be shown that (see here) the output is some algebraic sum of the inputs with both positive and negative coefficients:
(53) |
The main drawback of the differential amplifier is that its input
impedance () may not be high enough if the output impedance
of the source is high. To overcome this problem, two non-inverting
amplifiers with high input resistance are used each for one of the
two inputs to the differential amplifier. The resulting circuit is
called the instrumentation amplifier.
Recall that the output resistance of a non-inverting amplifier is very low, its output voltage will not affected by the load circuit, here the differential amplifier of which the input resistance is not very high. Therefore the outputs of the two non-inverters in the first stage of the instrumentation amplifier are:
(54) |
(55) |
Alternatively, we consider the current going from to
through
,
,
and
:
(56) |
(57) |
(58) |
(59) |
Without feedback, the output of an op-amp is
. As
is
large,
is saturated, equal to either the positive or the negative
voltage supply, depending on whether or not
is greater than
.
When an input of any waveform is compared with a reference voltage, the
output is a square wave:
These two possible outputs, positive and negative, can be treated as “1”
and “0” of the binary system. The figure shows an A/D converter built by
three op-amps to measure voltage from 0 to 3 volts with resolution
1 V.
Due to the voltage divider, the input voltages to the three op-amps are, respectively, 2.5V, 1.5V and 0.5V. The output of these op-amps are listed below for each of the input voltage levels. A digital logic circuit (a decoder) can convert the 3-bit output of the op-amps to the 2-bit binary representation.
(60) |
Integrator
In time domain, as
and
, we have (KCL)
(61) |
(62) |
Differentiator
If we swap the resistor and the capacitor, we get in time domain:
(63) |
(64) |
A proportional-integral-derivative (PID) controller can be implemented as shown. The output of the circuit is a linear combination of the signal together with its integral and derivative:
(65) |
Assuming
, we can show that the output current
through the load
is a constant determined by the input
voltages
and
, as well as the circuit parameters
(see here):
(66) |
Based on the relationship between the current through and voltage across a diode and the virtual ground assumption, we can show that the output voltage of the exponential amplifier (left) is approximately an exponential function of the input voltage, and the output voltage of the logarithmic amplifier (right) is approximately a logarithmic function of the input voltage:
(67) |
Many op-amp circuits practically used can be found here.