Power Supplies

Saturation Effects

Slew Rate Limitations

        Operational amplifiers are "active" devices.  They can produce signals that will cause more energy to be dissipated than the input signal provides.  (Amplifiers will almost always do that!)  Energy doesn't come out of the air.  It has to be provided.  That means that the operational amplifier must be connected to a power supply that will provide the energy that it puts into circuitry connected to the amplifier output.

        Consider an inverting amplifier with a gain of -10.  Assume that the input to the circuit is a signal with an amplitude of 1.4v.  You would expect the output of the amplifier to be a sinusoidal signal with an amplitude of 14 because the amplitude of the input is 1.4v and the gain is -10.  However, if you take saturation into account, you will get a signal that is "flattened" at the top and bottom.

        In this example, the sinusoidal signal saturates at +10.5 and -10.5 volts.  That's about what you would get with an amplifier using power supplies of +12v and -12v.  The amplifier usually saturates within a volt or so of the power supply voltages, and you can't get a larger output from an operational amplifier than the supply voltage.

        In the "oscilloscope" presentation below you can change the amplitude of the sinusoidal signal and see what saturation does to a sinusoidal signal.  And, later, you can check this in the lab.

        In our analysis of operational amplifier circuits we have assumed that the output of the operational amplifier can change instantaneously.  That's not what happens.  The rate of change of the output of an operational amplifier is limited.  For a 741 style operational amplifier, the limit could be 0.5 volts/microsecond.  In other words, if you want to change the output voltage from zero (0) volts to ten (10) volts, it would take twenty (20) microseconds.  Twenty microseconds may sound like a short time, but in many applications it isn't.

        Here is a simulation that shows what can happen when we try to produce a short square pulse.  Here the desired output is 10v for 40msec, and then 0v for 40msec.  However, the output of the amplifier cannot jump instantaneously to 10v.  Instead, it "ramps up" to 10v at a rate of 0.5 volts/microsecond, and that takes 20msec.  That's half of the time it should be up at 10v.  Then, when the pulse should go to zero, it ramps down at the same rate.  (NOTE:  In the simulation everything has been slowed down considerably.)