Professional Documents
Culture Documents
Op-amp circuits are often designed and implemented for signal differentiation and integration.
Until recently (before computer-based control), control algorithms (such as PID) containing
differentials and integrals were implemented in discrete circuit components. Differentiation is
also useful for obtaining velocity measurements from a signal representing a position or
determining a signal's frequency (recall the amplitude of the time derivative of a sinusoid is
scaled by its frequency). Figure 1 below shows an ideal op-amp integrator and differentiator with
input-output relationships that are theoretically correct, but have practical implementation issues
discussed below. In this lab, practically realizable differentiators and integrators will be built
using op-amps, resistors and capacitors.
1. The Differentiator
The ideal differentiator is inherently unstable in practice due to the presence of some high
frequency noise in every electronic system. An ideal differentiator would amplify this
small noise. For instance, if vnoise = Asin(t) is differentiated, the output would be vout =
Acos(t). Even if A = 1V, when = 210MHz) vout would have an amplitude of 63V!
To circumvent this problem, it is traditional to include a series resistor at the input and a
parallel capacitor across the feedback resistor as shown in figure 2, converting the
differentiator to an integrator at high frequencies for filtering.
Wire up the practical op-amp differentiator shown in figure 2 using your op-amp of
choice (741 or 356).
Drive it (via vin(t)) with a 1kHz sine wave, a 1kHz square wave, and a 1kHz triangle
wave. For each input signal, sketch the input and output waveforms.
Are the output waveforms and their amplitudes what you would expect, i.e., does the
circuit differentiate the input signal?
2. The Integrator
Op-amps allow you to make nearly perfect integrators such as the practical integrator
shown in figure 3. The circuit incorporates a large resistor in parallel with the feedback
capacitor. This is necessary because real op-amps have a small current flowing at their
input terminals called the "bias current". This current is typically a few nanoamps, and is
neglected in many circuits where the currents of interest are in the microamp to milliamp
range. However, if you apply a nanoamp current to a 0.1F capacitor, it won't take long
until it charges and becomes effectively an open circuit not allowing any current to flow!
The feedback resistor gives a path for the bias current to flow. The effect of the resistor
on the response is negligible at all but the lowest frequencies.
Drive the input (via vin(t)) with a 500Hz square wave of 2 V p-p amplitude. Sketch the
input and output waveforms.
Calculate the expected output waveform via integration using the circuit component
values and compare to the experimental waveform.
Does the amplitude of the output waveform agree with what it should be from the circuit
values?
In the DC condition Cf offers infinite resistance and so the integrator circuit will be like
an inverting opamp amplifier with infinite feedback resistance (Rf = ). The equation for
the voltage gain (A) of an opamp amplifier in inverting mode is A = -(Rf/R1).
Substituting Rf= in the present scenario we get A=. Therefore the small input offset
voltage will get amplified by this factor and there will be an error voltage at the output.
This problem can be solved by adding a feedback resistor Rf parallel to Cf as shown in
fig 4 shown below.