The MOSFET Amplifier Circuit Heart of Modern Electronics

From the micro-watt signals in your smartphone's radio to kilowatt audio systems and GHz-range processors, the Metal-Oxide-Semiconductor Field-Effect Transistor (mosfet) amplifier is a cornerstone of contemporary electronics. Its unparalleled combination of high input impedance, voltage-controlled operation, scalability, and efficiency has rendered it indispensable across analog, digital, and mixed-signal domains. This article delves into the operational principles, key configurations, design considerations, and pervasive applications of MOSFET amplifier circuits.

I. The MOSFET: Foundation of Amplification

Understanding the amplifier begins with the device itself. The MOSFET is a four-terminal device:

● Gate (G): Electrically isolated from the channel by a thin oxide layer (SiO₂). Acts as the control terminal. A voltage applied between Gate and Source modulates the conductivity of the channel.

● Source (S): The terminal through which majority carriers (electrons in nMOS, holes in pMOS) enter the channel.

● Drain (D): The terminal through which majority carriers leave the channel.

● Body/Bulk/Substrate (B): The physical substrate upon which the device is built. Typically connected to the most negative (for nMOS) or most positive (for pMOS) supply voltage to reverse-bias the source/drain junctions.

 

Key Operating Regions:

Cutoff (Subthreshold): V_GS < V_TH (Threshold Voltage). No significant channel forms. I_D ≈ 0.

Triode (Linear/Ohmic): V_GS > V_TH and V_DS < (V_GS - V_TH). A channel forms, and I_D depends strongly on both V_GS and V_DS. The device behaves like a voltage-controlled resistor.

Saturation (Active/Pinch-off): V_GS > V_TH and V_DS ≥ (V_GS - V_TH). The channel pinches off near the drain. I_D becomes primarily dependent on V_GS and relatively independent of V_DS (for long-channel devices). This is the primary region for amplification, analogous to the active region in BJTs. The drain current is approximated by: I_D = (1/2) * μ_n * C_ox * (W/L) * (V_GS - V_TH)^2 * (1 + λ * V_DS) Where:

μ_n: Electron mobility (nMOS)

C_ox: Gate oxide capacitance per unit area

W: Channel width

L: Channel length

V_GS: Gate-to-Source voltage

V_TH: Threshold Voltage

λ: Channel-length modulation parameter (accounts for the slight increase in I_D with V_DS)

 

Crucial Parameters for Amplifiers:

Transconductance (g_m): The most critical parameter for amplification. It defines the change in drain current for a change in gate-source voltage: g_m = ∂I_D / ∂V_GS. In saturation: g_m = μ_n * C_ox * (W/L) * (V_GS - V_TH) = √(2 * μ_n * C_ox * (W/L) * I_D) High g_m means high gain.

Output Resistance (r_o): Represents the small-signal resistance seen looking into the drain terminal when the gate is AC grounded. Caused by channel-length modulation: r_o ≈ 1 / (λ * I_D). Higher r_o is desirable for higher voltage gain.

Intrinsic Voltage Gain (A_V0): The maximum possible voltage gain from a single transistor in saturation: A_V0 = g_m * r_o. This is a fundamental figure of merit for the device technology.

Input Resistance (R_in): Extremely high (typically > 10^12 Ω for DC) due to the insulating oxide layer. Allows easy interfacing with high-impedance sources like sensors or previous amplifier stages without significant loading.

Gate Capacitances (C_gs, C_gd, C_gb): Parasitic capacitances between gate and source, gate and drain, and gate and body. These are critical determinants of high-frequency performance and stability.

 

II. Basic MOSFET Amplifier Configurations

Like their BJT counterparts, MOSFET amplifiers are categorized based on which terminal is common to both input and output signals.

Common-Source (CS) Amplifier: The Workhorse

Configuration: Source terminal is AC grounded (often via a large bypass capacitor). Input signal (V_in) applied to Gate. Output signal (V_out) taken from Drain.

Operation: A small positive change in V_GS increases I_D. This increased current flowing through the drain load resistor (R_D) causes a larger voltage drop across it, resulting in a negative change in V_D (V_out). Thus, the CS stage provides high voltage gain with a 180° phase inversion.

DC Biasing: Essential to establish the Q-point (quiescent operating point) in the saturation region. Common methods include:

Fixed Bias: Simple but unstable; V_GS fixed by a voltage divider. Highly sensitive to V_TH variations.

Self-Bias (Source Degeneration): A resistor (R_S) is placed between source and ground. The voltage drop across R_S (I_D * R_S) provides negative feedback, stabilizing the Q-point against variations in V_TH and temperature. Requires a bypass capacitor (C_S) across R_S for AC signals to maintain full gain.

Current Mirror Bias: Uses a reference current and a current mirror to set I_D precisely, independent of V_TH. Common in IC design.

AC Analysis & Gain:

With Ideal Bypassing (C_S present): Voltage Gain A_V = V_out / V_in = -g_m * (r_o || R_D || R_L). The gain is maximized by large g_m and large effective load resistance.

Without Bypassing (C_S absent): Source resistor R_S provides local negative feedback, reducing gain but improving linearity, bandwidth, and input impedance. Gain becomes A_V ≈ - (g_m * R_D) / (1 + g_m * R_S) (for r_o >> R_D).

Characteristics:

High Voltage Gain (moderate to high)

High Input Impedance (very high)

Moderate Output Impedance (≈ R_D || r_o)

Phase Inversion (180°)

Applications: Voltage amplification stages in audio preamps, sensor interfaces, operational amplifier input stages, general-purpose gain blocks.

Common-Drain (CD) Amplifier / Source Follower

Configuration: Drain terminal is AC grounded (usually connected to V_DD). Input signal (V_in) applied to Gate. Output signal (V_out) taken from Source.

Operation: A positive change in V_GS increases I_D. This increases the voltage drop across the source resistor (R_S), raising V_S (V_out). The output follows the input with a slight offset (≈ V_GS). Provides voltage gain slightly less than 1 (≈ 0.9 - 0.99), no phase shift, high input impedance, and low output impedance.

Gain: A_V = V_out / V_in = (g_m * (r_o || R_S || R_L)) / (1 + g_m * (r_o || R_S || R_L)) ≈ 1 (if g_m * R_effective >> 1). The body effect (if body not connected to source) reduces gain slightly.

Output Impedance (R_out): R_out = (1 / g_m) || r_o || R_S ≈ 1 / g_m (if r_o and R_S are large). This low output impedance is key.

Characteristics:

Voltage Gain ≈ 1 (Unity)

Very High Input Impedance

Very Low Output Impedance

No Phase Shift

Applications: Output buffer stages (to drive low-impedance loads like cables or speakers without loading the previous stage), level shifting, impedance transformation.

Common-Gate (CG) Amplifier

Configuration: Gate terminal is AC grounded. Input signal (V_in) applied to Source. Output signal (V_out) taken from Drain.

Operation: A positive change in V_in (applied to the Source) effectively decreases V_GS (since Gate is fixed), reducing I_D. The reduced current through R_D causes less voltage drop, resulting in a positive change in V_D (V_out). Provides moderate voltage gain, low input impedance, high output impedance, and no phase shift.

Gain: A_V = V_out / V_in = g_m * (r_o || R_D || R_L). Mathematically similar to CS gain but positive.

Input Impedance (R_in): R_in ≈ 1 / g_m (low). This is a defining characteristic.

Characteristics:

Moderate Voltage Gain

Low Input Impedance (≈ 1/g_m)

High Output Impedance (≈ R_D || r_o)

No Phase Shift

Applications: Current buffer (accepts low-impedance input current), high-frequency amplification (minimizes Miller effect), cascode configurations, input stages for low-impedance sources (e.g., some RF inputs, photodiodes).

 

III. Advanced Configurations and Techniques

1.The Cascode Amplifier:

Combines a CS stage (Q1) feeding into a CG stage (Q2). The input is applied to the gate of Q1; the output is taken from the drain of Q2. Q2's gate is held at a fixed DC bias.

Advantages:

Very High Voltage Gain: Approaching the intrinsic gain (g_m * r_o) of a single device squared (A_v ≈ (g_m1 * r_o1) * (g_m2 * r_o2)).

High Output Impedance: ≈ g_m2 * r_o2 * r_o1 (very high).

Improved Bandwidth: The CG stage (Q2) acts as a current buffer, isolating the input (Q1 drain) from the high-capacitance load. This drastically reduces the Miller effect capacitance (C_gd) multiplication on Q1, significantly extending bandwidth compared to a single CS stage.

Improved PSRR: Better power supply rejection.

Applications: High-gain, wideband amplifiers (e.g., RF front-ends, IF amplifiers, op-amp gain stages), high-impedance current sources.

2.Differential Amplifiers:

Uses a matched pair of MOSFETs (Q1, Q2) with sources connected together to a constant current source (I_SS). Inputs are applied differentially (V_in1, V_in2) to the two gates. Outputs can be taken single-endedly (from one drain) or differentially (between the two drains).

Operation: The tail current source forces the sum of I_D1 and I_D2 to be constant. A differential input voltage steers this current towards one transistor or the other.

Key Metrics:

Differential Gain (A_d): Gain for the difference signal: A_d = g_m * (r_o || R_D) (for single-ended output, half this). High.

Common-Mode Gain (A_cm): Gain for signals common to both inputs. Ideally zero. A_cm ≈ -R_D / (2 * r_oss), where r_oss is the output resistance of the tail current source. Low A_cm is crucial.

Common-Mode Rejection Ratio (CMRR): CMRR = |A_d / A_cm|. Should be very high (60-100 dB+). High CMRR rejects noise coupled equally onto both inputs (common-mode noise).

Input Common-Mode Range (ICMR): The range of DC input voltages over which the differential pair remains in saturation.

Applications: The fundamental input stage of nearly all operational amplifiers, instrumentation amplifiers, comparators, communication receivers (mixers), high-precision signal processing. Enables noise rejection and DC amplification.

3.Current Mirrors:

While not amplifiers themselves, current mirrors are vital for biasing MOSFET amplifiers in ICs. A reference current (I_ref) is set (e.g., using a resistor), and the mirror "copies" this current (or a scaled version) to other branches. Basic nMOS mirror consists of two matched transistors: Q_ref (diode-connected) and Q_copy. I_copy = (W/L)_copy / (W/L)_ref * I_ref (ignoring channel-length modulation). Cascode mirrors offer higher output resistance. Essential for setting precise, stable bias currents independently of power supply and threshold voltage variations.

 

IV. Critical Design Considerations

Biasing Stability: Establishing a stable Q-point in saturation is paramount. Self-bias and current mirror biasing offer significant advantages over fixed bias in terms of tolerance to device parameter variations (V_TH, μ_n, C_ox) and temperature. Temperature affects V_TH (decreases with temp) and mobility (decreases with temp), impacting I_D and g_m.

Gain vs. Bandwidth Trade-off (Miller Effect): This is a fundamental limitation, especially in CS stages. The gate-drain capacitance (C_gd) appears multiplied by (1 + |A_v|) at the input node due to the inverting gain. This "Miller capacitance" (C_Miller = C_gd * (1 + |A_v|)) significantly increases the total input capacitance (C_in ≈ C_gs + C_Miller), reducing bandwidth (f_max ≈ 1/(2π * R_s * C_in)). Techniques to mitigate:

● Cascode Configuration: As described earlier, isolates C_gd of the input device from the high-gain node.

● Source Degeneration (Unbypassed R_S): Reduces gain (A_v), thereby reducing C_Miller, but also reduces gain.

● Inductive Peaking/Shunt Peaking: Adding inductance to resonate with capacitance and extend bandwidth near a specific frequency. Common in RF amplifiers.

Frequency Response: Beyond the Miller effect, the intrinsic device capacitances (C_gs, C_gd, C_db, C_sb) and load capacitance (C_L) form low-pass filters with circuit resistances, limiting high-frequency gain. The dominant pole is usually at the highest impedance node (often the output node). Careful analysis (e.g., using zero-value time constants or simulation) is needed to predict bandwidth and phase margin (for stability).

Linearity and Distortion: MOSFETs operating in saturation are inherently square-law devices (I_D ∝ (V_GS - V_TH)^2). This non-linearity causes harmonic distortion (HD2, HD3) and intermodulation distortion (IMD) when amplifying large signals or complex waveforms. Techniques to improve linearity:

● Source Degeneration: Introduces local negative feedback, linearizing the transconductance (g_m becomes ≈ 1 / R_S for large g_m * R_S).

● Large Overdrive Voltage (V_GS - V_TH): Moves operation into a more linear region of the transconductance curve relative to the signal swing.

● Differential Pair: Inherently cancels even-order harmonics when used with differential signaling.

● Feedback: Global negative feedback (as used in op-amps) is the most powerful method for reducing distortion.

Power Consumption: Crucial for portable and high-density ICs. Static power (I_D * V_DS at Q-point) and dynamic power (C * V^2 * f, from charging/discharging capacitances) must be minimized. Techniques include subthreshold operation (very low I_D, low speed), power gating, and efficient class designs (e.g., Class D switching amplifiers).

Noise Performance: MOSFETs introduce several noise sources:

● Thermal Noise: From channel resistance (i_nd^2 = 4kT * γ * g_m * Δf, γ ≈ 2/3 for long-channel, higher for short).

● Flicker Noise (1/f Noise): Dominant at low frequencies, caused by carrier trapping/release at the Si-SiO₂ interface (i_nd^2 ∝ 1/f). Proportional to 1/(W * L * C_ox). PMOS often has lower 1/f noise than NMOS. Minimizing noise involves maximizing g_m for the bias current (large W/L), using larger devices (reduces 1/f), choosing PMOS for input stages where low-frequency noise is critical, and employing correlated double sampling (CDS) techniques.

Process Variation: In IC design, parameters like V_TH, μ_n, C_ox, L, W vary across the wafer and between fabrication runs. Designs must be robust (e.g., using feedback, current mirrors, differential structures) or include trimming capabilities to meet specifications despite variations.

Body Effect: When the source and body are not at the same potential (common in ICs where all NMOS bodies are tied to GND), the threshold voltage V_TH increases: V_TH = V_TH0 + γ * (√|2φ_F + V_SB| - √|2φ_F|), where γ is the body-effect coefficient and V_SB is the source-to-body voltage. This reduces g_m and gain in CS and SF stages, and introduces signal-dependent shifts. Connecting the source to the body (if possible in layout) eliminates this.

V. Applications Across the Spectrum

1.Operational Amplifiers (Op-Amps):

The input stage is almost universally a MOSFET differential pair (for high input impedance and good CMRR), followed by cascode gain stages (for high gain and bandwidth), and a source follower output stage (for low output impedance). CMOS op-amps dominate due to low power, high integration, and rail-to-rail capability.

2.Radio Frequency (RF) Amplifiers:

● Low-Noise Amplifiers (LNAs): Critical first stage in receivers. Designed for maximum power gain (or minimum noise figure, NF) while matching to the antenna/feedline impedance (50Ω). Common-source or cascode configurations with inductive source degeneration for simultaneous noise and input matching are prevalent. Extremely high-frequency designs use distributed amplifiers.

● Power Amplifiers (PAs): Deliver high power to antennas. Classes A, AB, B (linear) and D, E, F (switching) are implemented using power MOSFETs (LDMOS, GaN-on-Si) for efficiency at GHz frequencies (cellular base stations, radar).

3.Digital Circuits:

While not "amplifiers" in the traditional analog sense, CMOS inverters and logic gates are essentially highly non-linear, high-gain voltage amplifiers switching between saturation and cutoff. The speed and power consumption of digital systems fundamentally depend on MOSFET amplifier characteristics (g_m, C, V_TH).

4.Audio Amplifiers:

Preamplifiers: High-gain, low-noise stages (often CS or differential) for weak signals from microphones or instruments.

Power Amplifiers: High-power output stages driving speakers. MOSFETs (especially lateral DMOS) are favored for their "softer" clipping characteristics compared to BJTs and simpler drive requirements. Class D (switching) amplifiers using MOSFETs achieve very high efficiency (>90%).

5.Sensor Interfaces:

High-input-impedance MOSFET amplifiers (source followers, instrumentation amps built with op-amps) are essential for interfacing with high-impedance sensors like piezoelectric transducers, pH electrodes, and photodiodes without loading them.

6.Switched-Capacitor Circuits:

Found in filters, ADCs, DACs. Rely on MOSFETs as switches controlled by clock signals and as amplifiers (usually op-amps) in integrators. The precision depends on MOSFET on-resistance and charge injection.

7.Memory Circuits:

dram cells use a MOSFET as an access transistor. Flash memory uses floating-gate MOSFETs for storage. Sense amplifiers (differential amps) are critical for reading the small signals from memory cells.

 

VI. Conclusion

The MOSFET amplifier, born from the fundamental physics of the silicon-silicon dioxide interface and field-effect control, has evolved into the dominant force in electronic amplification. Its journey from discrete power stages to billions of nanoscale devices integrated onto a single chip underscores its versatility and scalability. The core configurations – Common-Source, Common-Drain, and Common-Gate – provide the essential building blocks, while advanced techniques like cascoding and differential pairing address critical challenges of gain, bandwidth, noise, and linearity.

Designing effective MOSFET amplifiers requires careful navigation of complex trade-offs: gain versus bandwidth, linearity versus power, noise versus speed, and performance versus robustness to manufacturing variations. The relentless drive for smaller, faster, lower-power, and higher-frequency electronics continuously pushes MOSFET technology and circuit design to new frontiers, with FinFETs, GaAsFET, and novel materials like SiGe and III-V compounds extending Moore's Law and enabling applications from ubiquitous IoT devices to exascale computing and next-generation wireless communication.

From amplifying the faint whispers of distant stars in radio telescopes to driving the thunderous output of concert halls, the MOSFET amplifier remains an indispensable and dynamically evolving engine of the electronic age. Its mastery is fundamental for anyone shaping the future of electronics.

Recommended articles related to this topic：

Can TVS DIODES and ESD diodes be interchanged? unikeyic, come and educate!