Differential Amplifier Fundamentals

Differential amplifiers are the workhorses of precision analog electronics, enabling systems to extract weak signals buried in noisy environments. By amplifying the difference between two input voltages while rejecting any interference common to both, they form the critical input stage for operational amplifiers, data acquisition systems, and sensitive measurement equipment. Understanding their operation is essential for designing robust circuits that perform accurately in the real world, where electrical noise is always present.

The Core Principle: Amplifying Difference, Rejecting Commonality

At its heart, a differential amplifier is a circuit that amplifies the voltage difference between two input terminals. The key metric for any differential amplifier is its ability to perform this task while ignoring signals that appear identically on both inputs. This unwanted signal is called the common-mode voltage. In practice, common-mode voltage often comes from environmental noise, ground loops, or interference picked up by long cables. A perfect differential amplifier would have infinite common-mode rejection, outputting a signal based solely on the difference between its inputs. Real-world circuits strive to approximate this ideal, making the differential amplifier indispensable for applications ranging from audio processing to biomedical sensor interfaces.

Anatomy of the Basic Differential Pair

The most fundamental implementation of a differential amplifier is the differential pair, also known as a long-tailed pair. This circuit uses two matched transistors, typically BJTs or MOSFETs, whose characteristics are as identical as possible. These transistors share a common current source connected to their emitters (for BJTs) or sources (for MOSFETs). This current source is crucial; it provides a fixed total bias current, forcing the transistors to operate in a push-pull manner.

When an input voltage difference is applied, one transistor conducts more of the shared current while the other conducts less. This creates a proportional difference in their collector or drain currents, which is then converted into an output voltage. If the same voltage (a common-mode signal) is applied to both inputs, the current source attempts to keep the total current constant. Ideally, the current split between the two transistors doesn't change, resulting in no change in the output voltage. This is the essence of common-mode rejection. The quality of the current source and the matching of the transistors directly determine how well the circuit rejects common noise.

Quantifying Performance: Gain and CMRR

A differential amplifier has two distinct gains. The differential gain ($A_d$) is the amplification factor for the voltage difference between the two inputs. If $V_1$ and $V_2$ are the input voltages, the differential input is $V_d=V_1-V_2$. The output due to this difference is $V_{out,d}=A_d\cdot V_d$.

Conversely, the common-mode gain ($A_{cm}$) is the (undesired) amplification of the average voltage applied to both inputs. The common-mode input is $V_{cm}=(V_1+V_2)/2$. The output due to this common signal is $V_{out,cm}=A_{cm}\cdot V_{cm}$.

The ratio of these two gains defines the common-mode rejection ratio (CMRR), which quantifies the circuit's ability to reject common signals. It is typically expressed in decibels (dB): $$\text{CMRR (dB)}=20{\log }_{10}\left(\frac{A_d}{A_{cm}}\right)$$ A higher CMRR indicates better rejection. For example, a CMRR of 100 dB means the differential signal is amplified 100,000 times more effectively than the common-mode signal. In a well-designed discrete differential pair, CMRR is primarily limited by the mismatch between the two transistors and the finite output impedance of the common current source.

From Pairs to Integrated Circuits: The Op-Amp Connection

The differential pair is not just a standalone circuit; it is the foundational input stage of virtually all operational amplifiers (op-amps). In an op-amp IC, the differential pair provides the high input impedance and excellent common-mode rejection that make op-amps so versatile. The output of the differential pair is then fed into additional gain stages and a output buffer to create the high-gain, differential-to-single-ended amplifier we use as a building block.

This architecture is extended further in instrumentation amplifiers, which are specialized for sensor measurement. An instrumentation amplifier typically uses two or three op-amps configured to provide extremely high input impedance, very high CMRR, and a gain that can be set with a single resistor. The input stage of an instrumentation amplifier is a precision differential pair, often with laser-trimmed resistors to ensure optimal matching and maximize CMRR, which can exceed 120 dB. This makes them ideal for amplifying signals from strain gauges, thermocouples, and other low-level sources in electrically noisy industrial environments.

Common Pitfalls

1. Ignoring Transistor Mismatch: Assuming the two transistors in a differential pair are perfectly identical is a major error. Even small variations in their base-emitter voltage ($V_{BE}$) or beta ($\beta$) will degrade CMRR significantly. This is why integrated circuits use careful layout techniques and monolithic transistor pairs, and discrete designs may require selected components or trimming potentiometers.

• Correction: Always account for mismatch in your design calculations. For critical applications, use matched transistor arrays or integrated differential amplifier ICs where the matching is done for you.

1. Using a Poor Common Current Source: Replacing the current source with a simple resistor is a common simplification that severely compromises performance. A resistor has a low incremental resistance, allowing the common-mode signal to vary the tail current, which in turn modulates the transistor biases and creates a large common-mode output.

• Correction: Implement a proper active current source, such as a transistor with a bias voltage or a dedicated current source IC. This provides a high output impedance, which is essential for maintaining a constant tail current under common-mode voltage shifts.

1. Overlooking the Common-Mode Input Range: Every differential amplifier has limits on the voltage levels it can accept at its inputs before the transistors saturate or the current source ceases to operate properly. Applying inputs outside this common-mode range will cause the circuit to fail, distort the output, or lose its common-mode rejection entirely.

• Correction: Always consult the datasheet for integrated amplifiers or perform a DC bias analysis for discrete designs to ensure your expected input voltages fall within the specified common-mode range.

1. Neglecting Impedance Balance: The CMRR of a differential amplifier depends not only on the circuit itself but also on the source impedances connected to its inputs. If the impedance looking out from each input terminal is not equal, any common-mode noise voltage will create an unbalanced voltage divider, converting common-mode noise into a differential signal that will be amplified.

• Correction: Strive for symmetric source impedances. In sensitive applications, such as with instrumentation amplifiers, use shielded cables and ensure both signal lines have identical resistance and capacitance to ground.

Summary

• The differential amplifier is a circuit designed to amplify the voltage difference between two inputs while rejecting signals common to both, a capability critical for noise immunity in analog systems.
• Its fundamental building block is the differential pair, which uses two matched transistors biased by a common current source to achieve this differential amplification and common-mode rejection.
• Performance is measured by the common-mode rejection ratio (CMRR), the ratio of differential gain to common-mode gain, which quantifies how effectively the circuit suppresses unwanted common-mode signals.
• The differential pair forms the essential input stage of operational amplifiers and is the core of precision instrumentation amplifiers, enabling high-accuracy signal conditioning from sensors and other low-level sources.
• Successful implementation requires attention to component matching, the use of a high-impedance current source, and careful consideration of the common-mode input range and external impedance balance.