The term instrumentation amplifier is often misused, referring to the application rather than the architecture of the device. Historically, any amplifier that was considered precision (i.e., implemented some sort of input offset correction) was thought to be an “instrumentation amplifier,” since it was designed for use in measurement systems. Instrumentation amplifiers, or INAs, are related to operational amplifiers (op amps), in that they are based on the same basic building blocks. But an INA is a specialized device, designed for a specific function, as opposed to a fundamental building block. In this regard, instrumentation amplifiers are not op amps, in that they are designed to function differently.
Perhaps the most notable difference between an INA and an op amp, in terms of usage is the lack of a feedback loop. Op amps can be configured to perform a wide variety of functions, including inverting gain, non-inverting gain, voltage follower, integrator, low-pass filter, high-pass filter and many more. In all cases, the user is providing a feedback loop from the output of the op amp to the input, and that feedback loop determines the function of the amplifier circuit. This flexibility is why op amps are so prolific in a wide variety of applications. An INA, on the other hand, has this feedback internally, so there isn’t an external feedback to the input pins. For an INA, the configuration is limited to one or two external resistors, or perhaps a programmable register, to set the gain of the amplifier.
INAs are specifically designed and used for their differential-gain and common-mode-rejection capabilities. The instrumentation amplifier will amplify the difference between the inverting and non-inverting inputs while rejecting any signal that is common to both inputs, resulting in no common-mode component being present at the output of the INA. An op amp that is configured for gain (either inverting or non-inverting) will amplify the input signal by the set closed-loop gain, but the common-mode signal will remain at the output. The difference in gain between the signal of interest and the common-mode signal results in a reduction of common mode (as a percentage of the differential signal), but the common mode is still present at the output of the op amp, which limits the dynamic range of the output.
As mentioned, INAs are used to extract a small signal in the presence of a large common mode, but this common-mode component can take many forms. When using a sensor in a Wheatstone bridge configuration (which we will explore later), there is a large DC voltage that is common to both inputs. However, interference signals can take many forms; one common source is 50 or 60 Hz interference from the power lines, not to mention the harmonics. This time-varying error source often fluctuates greatly across frequency as well, making it extremely difficult to compensate for at the output of the instrumentation amplifier. These variances make specifying common-mode rejection, not only at DC but across a range of frequencies, important.