Common drain
The common drain amplifier, also known as the source follower, is a fundamental configuration of a metal-oxide-semiconductor field-effect transistor (MOSFET) amplifier in which the input signal is applied to the gate terminal, the output voltage is taken from the source terminal, and the drain terminal is connected to a common point, typically AC ground through a supply voltage.[1][2] This setup operates the MOSFET in saturation mode, with the source voltage closely following the gate voltage due to the device's transconductance properties.[1]
Key characteristics of the common drain amplifier include a voltage gain of approximately unity (A_v ≈ 1), which can be expressed as A_v = g_m / (g_m + 1 / (r_o ∥ r_{oc})), where g_m is the transconductance, r_o is the output resistance of the MOSFET, and r_{oc} is the resistance at the coupling point; this near-unity gain ensures minimal signal attenuation.[1] It exhibits very high input impedance, effectively infinite at the gate due to the insulating oxide layer, often limited only by biasing resistors in the range of 1 kΩ to 2 MΩ, which minimizes loading on the preceding stage.[2][3] Conversely, the output impedance is low, approximately 1/g_m (typically tens to hundreds of ohms), enabling it to drive loads effectively without significant voltage drop.[1][2]
This configuration is particularly valued for its role as a voltage buffer in multistage amplifier circuits, where it isolates stages with differing impedance requirements, such as matching a high-impedance source to a low-impedance load, and provides signal isolation using a single supply voltage like +12 V.[1][3] In practice, an isolation resistor (e.g., 47 Ω) may be added at the output to handle capacitive loads, and the amplifier's open-loop voltage gain exceeds 0.4 V/V, comparable to its bipolar junction transistor counterpart, the common collector amplifier.[3] Its simplicity, requiring minimal components beyond biasing networks, makes it a staple in analog integrated circuits and discrete designs for applications like impedance matching and level shifting.[2]
Overview
Definition and principles
The common drain (CD) amplifier, also known as a source follower, is a field-effect transistor (FET) amplifier topology in which the drain terminal is connected to a fixed supply voltage, effectively acting as an AC ground, the input signal is applied to the gate, and the output is taken from the source terminal.[4][5] This configuration provides voltage buffering, where the output voltage closely tracks the input voltage applied at the gate.[6]
The core principles of the common drain amplifier revolve around its unity voltage gain, where the voltage gain A_v \approx 1, resulting in a non-inverting output that follows the input signal with minimal attenuation.[4] It exhibits very high input impedance, which minimizes loading of the preceding stage.[6] Additionally, the output impedance is low, approximately Z_{out} \approx \frac{1}{g_m} (where g_m is the transconductance of the FET), enabling it to drive subsequent low-impedance loads effectively without significant voltage drop.[6] These characteristics make it ideal for applications requiring impedance matching and signal isolation.[4]
Prerequisite to understanding the common drain amplifier is the basic operation of FETs, which are unipolar devices where current flow between the drain and source is controlled by an electric field established by the gate voltage. FETs can operate in depletion mode, where the channel conducts at zero gate voltage and conduction decreases with increasing reverse bias, or enhancement mode, where a channel forms only when forward bias is applied to turn the device on. They are further classified as n-channel, using electrons as majority carriers for faster operation, or p-channel, using holes, allowing complementary circuit designs.
The common drain configuration was first described in early transistor literature around the 1950s, serving as the FET analog to the bipolar junction transistor's emitter follower for similar buffering purposes.[4]
Historical context
The common drain configuration, also known as the source follower, emerged in the early 1950s alongside the theoretical development of the junction field-effect transistor (JFET) at Bell Laboratories. William Shockley proposed the unipolar field-effect transistor in 1952, explicitly describing a source follower arrangement analogous to the vacuum tube cathode follower, which offered high input impedance and low output impedance for buffering applications.[7] This theoretical foundation linked the FET's operation to established vacuum tube circuits, enabling voltage gain near unity while isolating stages. Practical realization followed shortly, as George C. Dacey and Ian M. Ross fabricated the first working JFETs in 1953, demonstrating their viability in amplifier configurations including the common drain, under Shockley's supervision.[8]
The configuration gained prominence in the MOSFET era beginning in the late 1950s, when Mohamed M. Atalla and Dawon Kahng invented the metal-oxide-semiconductor field-effect transistor (MOSFET) in 1959 at Bell Labs, fabricating the first device and demonstrating its use in field-effect amplifiers by 1960. This innovation facilitated the integration of common drain stages into early integrated circuits during the 1960s, where MOSFETs' scalability and compatibility with silicon processes proved advantageous for compact electronics. By the 1970s, the common drain saw widespread adoption in operational amplifiers and RF designs; for instance, Texas Instruments introduced the TL071 JFET-input op-amp in 1972, employing source follower input stages for low-noise, high-impedance applications in audio and instrumentation circuits.
Key milestones include the configuration's integration into very-large-scale integration (VLSI) during the 1980s, as CMOS technology matured, allowing common drain buffers to support high-density logic and analog interfaces in microprocessors and memory chips. In the post-2000 era, its relevance persists in low-power CMOS designs for mobile and IoT devices, where Atalla and Kahng's MOSFET advancements enable efficient voltage buffering with minimal energy consumption, as evidenced in modern submicron processes. Shockley's theoretical work, combined with Atalla and Kahng's practical MOSFET enabling, established the common drain as a foundational element in semiconductor electronics.
Circuit configuration
Basic schematic
The basic schematic of a common drain amplifier, also known as a source follower, features an n-channel enhancement-mode MOSFET as the active device, with the drain terminal connected directly to the positive supply voltage V_DD, which serves as the AC ground reference.[9] The input signal is applied to the gate terminal, typically through a coupling capacitor and a bias resistor network consisting of two resistors (R1 and R2) forming a voltage divider to set the DC operating point.[10] The output is taken from the source terminal, which is connected to ground through a source resistor R_S for DC bias and output swing, with the load resistor R_L often connected in parallel with R_S or AC-coupled from the source; a bypass capacitor may parallel the drain connection to V_DD to ensure AC grounding.[9]
Key components include the gate bias resistors, which provide a high-impedance DC path while blocking AC signals via coupling capacitors, ensuring the gate input remains isolated for signal integrity.[10] The source resistor R_S establishes the quiescent current and allows for voltage output at the source, enabling the follower action where the source voltage tracks the gate input.[9] In standard drawings, the n-channel MOSFET is symbolized by a gate line overlapping the channel between drain and source arrows indicating the n-type substrate, with signal flow annotated from gate input to source output using arrows or labels for clarity.[10]
Variants of the schematic include single-ended configurations for basic applications, though the core topology remains unchanged.[9] Implementations can use JFETs instead of MOSFETs, where the drain connects to V_DD, the gate receives the input via a high-value resistor R_G for self-bias, and the source outputs through R_S to ground, employing the standard JFET symbol with a channel bar and gate arrow pointing inward for n-channel devices.[11] As with MOSFETs, the JFET configuration leverages the gate, drain, and source terminals for input, AC ground, and output functions, respectively.
Biasing arrangements
In the common drain configuration, proper biasing establishes the DC operating point to ensure the transistor operates in the saturation region, where the drain-to-source voltage satisfies V_{DS} > V_{GS} - V_{th} for MOSFETs or the pinch-off condition for JFETs, enabling linear amplification without distortion.[12][9] This involves setting the gate-to-source voltage V_{GS} and drain current I_D to center the quiescent point within the device's characteristics, accounting for variations in threshold voltage and transconductance.
The self-bias technique, commonly used for JFETs and depletion-mode MOSFETs, employs a source resistor R_S to generate the required V_{GS}, with the gate connected to ground either directly or through a high-value resistor for AC grounding. For an n-channel JFET, the negative V_{GS} develops across R_S as I_D flows, stabilizing the bias against device variations; the quiescent current is determined by the intersection of the load line I_D = -V_{GS}/R_S with the transfer characteristic.[13][14] This method requires no additional DC supply but provides moderate thermal stability due to the negative temperature coefficient of the pinch-off voltage in JFETs, which helps prevent thermal runaway by reducing I_D as temperature rises.[13]
For enhanced precision, particularly in enhancement-mode MOSFETs where positive V_{GS} > V_{th} is needed, voltage divider bias uses resistors R_1 and R_2 connected to the supply rails to set the gate voltage V_G. The Thevenin equivalent voltage is V_{GG} = V_{DD} \frac{R_2}{R_1 + R_2}, and the loop equation yields I_D = \frac{V_{GG} - V_{GS}}{R_S}, where V_{GS} is solved iteratively from the device equation, such as the square-law model for MOSFETs.[9][12] This arrangement improves stability over self-bias by reducing sensitivity to V_{th} variations, though R_1 and R_2 must be large (e.g., >10 MΩ) to maintain high input impedance.[13]
Current source bias enhances stability by replacing R_S with a constant current sink at the source, setting I_D directly and minimizing the impact of supply variations or temperature changes. In this setup, a bias current I_{SUP} (e.g., 0.4 mA) flows through the device, with V_{GS} adjusted via the gate voltage to satisfy the saturation condition, often using an external current mirror or depletion-mode JFET as the sink.[13][12] This technique is ideal for source followers requiring low output impedance and is particularly effective for JFETs, where it further mitigates thermal runaway by clamping I_D independently of V_{GS} shifts.[14]
Key considerations include verifying the saturation region post-biasing to avoid triode operation, which degrades gain; for instance, V_{DS} = V_{DD} - I_D R_S > V_{GS} - V_{th} must hold.[9] In JFET circuits, additional measures like source degeneration via unbypassed R_S prevent thermal runaway by providing negative feedback that counteracts I_D increases with temperature.[13]
Analysis
Small-signal equivalent circuit
The small-signal equivalent circuit for the common drain (CD) amplifier, also known as a source follower, employs the hybrid-π model of the MOSFET to analyze AC performance under small-signal conditions. This linearized representation replaces the nonlinear MOSFET with linear elements valid for signal variations much smaller than the DC bias levels, facilitating analysis of voltage gain, input/output impedances, and frequency response.[12][15]
Key assumptions include operation in the saturation region around the DC quiescent point (Q-point), where the drain-source voltage exceeds the gate-source overdrive, and small-signal approximations that neglect higher-order nonlinearities. For simplicity in basic models, an infinite Early voltage is often assumed, implying no channel-length modulation and thus infinite output resistance; however, more accurate models incorporate finite output resistance to account for this effect.[12][16][15]
In the equivalent circuit, the drain terminal is AC-grounded, the input signal is applied to the gate, and the output is taken from the source. The MOSFET is modeled as follows:
- A voltage-controlled current source g_m v_{gs}, where g_m is the transconductance (g_m = 2 \sqrt{K I_D}, with K as the process transconductance parameter and I_D the DC drain current) and v_{gs} the small-signal gate-source voltage; this source connects between the source and drain (ground) nodes.[15][12]
- An output resistance r_o in parallel with the current source, from source to drain (ground), given by r_o = V_A / I_D (where V_A = 1/\lambda is the Early voltage and \lambda the channel-length modulation parameter) to model parasitic effects due to finite channel length.[15][16]
- For high-frequency analysis, gate-source capacitance C_{gs} (approximately \frac{2}{3} W L C_{ox}, with W and L as channel width and length, and C_{ox} the oxide capacitance per unit area) between gate and source, and gate-drain capacitance C_{gd} (typically small, 1–10 fF) between gate and drain (ground); these parasitics introduce poles and zeros affecting bandwidth.[15][12]
The circuit configuration appears as:
[Gate](/page/Gate) (input) ── v_gs ── C_gs ── [Source](/page/Source) (output)
│
C_gd
│
[Drain](/page/Drain) (ground)
│
┌────┴────┐
│ g_m v_gs │
│ || │
│ r_o │
└────┴────┘
│
Ground
[Gate](/page/Gate) (input) ── v_gs ── C_gs ── [Source](/page/Source) (output)
│
C_gd
│
[Drain](/page/Drain) (ground)
│
┌────┴────┐
│ g_m v_gs │
│ || │
│ r_o │
└────┴────┘
│
Ground
This model, valid when superimposed on the Q-point from biasing, enables subsequent AC analysis without deriving specific parameters here.[16][12]
Voltage gain derivation
The voltage gain of a common drain amplifier, also known as a source follower, is derived using the small-signal equivalent circuit, which models the MOSFET with transconductance g_m, gate-source voltage v_{gs}, and output resistance r_o.[12]
In the basic configuration, the drain is AC-grounded, the input signal v_{in} is applied to the gate, and the output v_{out} is taken from the source, which is connected to an effective load resistance R_S' (typically the source resistor R_S in parallel with any external load). The gate-source voltage relates to the input and output as v_{gs} = v_{in} - v_{out}. Applying Kirchhoff's current law at the source node, the controlled current source g_m v_{gs} flows through the effective load, yielding v_{out} = g_m v_{gs} R_S'. Substituting the expression for v_{gs} gives v_{out} = g_m (v_{in} - v_{out}) R_S', which rearranges to v_{out} (1 + g_m R_S') = g_m v_{in} R_S'. Thus, the low-frequency small-signal voltage gain is A_v = \frac{v_{out}}{v_{in}} = \frac{g_m R_S'}{1 + g_m R_S'}.[12][17]
For the ideal case, where channel-length modulation is neglected (i.e., r_o \to \infty) and g_m R_S' \gg 1, the gain approximates unity: A_v \approx 1. This near-unity gain, combined with high input impedance, makes the common drain suitable as a voltage buffer.[12]
To account for finite output resistance r_o, the effective load becomes R_S' = R_S \parallel r_o = \frac{R_S r_o}{R_S + r_o}, modifying the gain to A_v = \frac{g_m (R_S \parallel r_o)}{1 + g_m (R_S \parallel r_o)}. The presence of r_o slightly reduces the gain from the ideal value, particularly when r_o is comparable to R_S.[12][17]
A key aspect of the voltage gain analysis is the bootstrapping effect, where the source voltage closely tracks the gate voltage, reducing the effective voltage swing across the gate-source capacitance and thereby minimizing input capacitance variations that could otherwise degrade gain at higher frequencies.[18]
Impedance properties
The common drain amplifier exhibits a very high input impedance, primarily limited by the gate bias resistor R_G in parallel with the FET's gate, which draws negligible current due to its insulating oxide layer, resulting in values typically in the range of 1 MΩ to 10 MΩ, depending on the choice of biasing resistors R_G.[12][19] This high input impedance makes the configuration suitable for interfacing with high-impedance sources without significant loading.
The output impedance is low, approximated by Z_{out} \approx \frac{1}{g_m} \parallel R_S when unloaded, where g_m is the transconductance of the FET; for typical discrete or integrated FETs with g_m in the range of 1–10 mS, this yields values of 100–1000 \Omega. This low output impedance arises from the source follower action, where the FET actively drives the load, providing current buffering.
The Miller effect is significantly reduced in the common drain configuration because the voltage gain is approximately unity, minimizing the multiplication of the gate-drain capacitance C_{gd} at the input to roughly C_{gd} (1 - A_v) \approx 0.[22] This limited impact on input capacitance contributes to the amplifier's favorable high-frequency behavior at low frequencies.
To characterize the output impedance, the Thevenin equivalent circuit is used, looking backward into the source terminal of the FET with the input shorted for small-signal analysis.[9][19]
Bandwidth and stability
The frequency response of a common drain amplifier is primarily determined by the dominant pole arising from the gate-source capacitance C_{gs} and the effective output resistance at the source node. In the small-signal model, this pole dominates when other higher-frequency effects are negligible, leading to a -3 dB bandwidth approximated by f_{-3dB} \approx \frac{1 + g_m R_S'}{2\pi R_S' C_{gs}}, where R_S' represents the effective source resistance (often the parallel combination of the source load and $1/g_m) and g_m is the transconductance.[23][24] This approximation holds for configurations where the low-frequency voltage gain is near unity, emphasizing the role of C_{gs} in charging through the source resistance.
The gain-bandwidth product for the common drain amplifier approaches \frac{g_m}{2\pi C_{gs}}, akin to that of ideal unity-gain buffers, reflecting the inherent trade-off between transconductance and input capacitance in MOSFET devices.[24] This product remains relatively constant, allowing high bandwidth at low gain but limiting overall performance in high-capacitance scenarios.
Stability in common drain circuits can be compromised when the output is loaded with a capacitive load, potentially leading to oscillations due to the negative input impedance and feedback effects from parasitic capacitances like the nonlinear C_{gs}.[25] Compensation is often achieved through source degeneration, where a resistor in series with the source provides negative feedback to enhance stability and linearity, mitigating risks across a broad frequency range.[25]
At high frequencies, the gate-drain capacitance C_{gd} further limits bandwidth by introducing additional poles and feedforward paths that couple the input to the output, effectively reducing the overall frequency response.[5] This effect becomes prominent when C_{gd} interacts with the load, shorting signals and degrading the buffer's isolation.[23]
Applications
Buffer amplifier uses
The common drain amplifier serves as a voltage buffer, isolating high-impedance sources such as sensors from low-impedance loads while providing unity voltage gain to prevent signal attenuation.[12] This configuration ensures that the input signal at the gate is faithfully reproduced at the source output, with minimal loading effects on the preceding stage due to its inherently high input impedance.
In impedance matching applications, the common drain's high input impedance makes it suitable for interfacing with piezoelectric transducers, which generate high-impedance charge signals that require buffering to avoid loss.[26] Conversely, its low output impedance enables effective driving of low-impedance loads, such as coaxial cables, without significant voltage drop or distortion.[12] These impedance properties, as detailed in the performance characteristics, facilitate efficient signal transfer across mismatched interfaces.
Operating in Class A mode, the common drain amplifier achieves low distortion through constant bias current but is limited to a maximum efficiency of approximately 25% due to continuous power dissipation in the transistor.[27] This trade-off prioritizes linearity over power efficiency, making it ideal for precision applications where harmonic distortion must be minimized.
Representative uses include audio preamplifier stages, where it buffers microphone signals with high input impedance to preserve fidelity before amplification. In analog integrated circuits, it functions as a voltage level shifter, providing DC offset adjustment between stages while maintaining signal integrity.[28]
Specific circuit examples
One practical implementation of the common drain configuration is in audio buffer amplifiers. A MOSFET source follower circuit, using complementary N-channel and P-channel devices such as the 2SK1058/2SJ162 pair, operates as the output stage with gate resistors around 200 Ω. This design provides high current capability and low output impedance suitable for audio loads, powered by a ±55 V supply for high-fidelity applications.[29]
In RF applications, the common drain JFET serves as an impedance matcher in antenna preamplifiers, where its high input impedance couples to the high-impedance antenna while the output is tuned to 50 Ω for compatibility with receiver front-ends.[30]
The common drain topology is also employed in CMOS gate drivers for power MOSFETs, particularly in half-bridge configurations requiring high-side driving. Integrated drivers use a bootstrap circuit with a diode and capacitor (e.g., C_BST sized based on gate charge Q_G and on-time t_ON) to float the source follower stage, enabling gate voltages up to 15-20 V above the high-side source for efficient switching in motor control or DC-DC converters.[31]
Comparisons
With common source
The common drain (CD) amplifier configuration contrasts with the common source (CS) in its fundamental performance metrics, particularly voltage gain and impedance characteristics. In the CS setup, the voltage gain is inherently high and inverting, expressed as A_v = -g_m R_D, where g_m is the transconductance and R_D is the drain resistance, enabling significant signal amplification.[32] By comparison, the CD amplifier delivers near-unity voltage gain, approximately A_v \approx 1 (non-inverting), as the output at the source closely follows the input at the gate, prioritizing signal fidelity over amplification.[17] This unity gain in CD arises from its source follower topology, which lacks the drain load resistor's amplifying effect present in CS.[5]
Impedance properties further highlight these differences, with both configurations sharing high input impedance due to the gate's insulating nature, but diverging at the output. The CS amplifier presents high output impedance, primarily R_o = R_D, which suits applications needing minimal loading on prior stages but can limit drive capability for low-impedance loads.[32] In contrast, the CD offers low output impedance, approximately R_o = 1/g_m, facilitating efficient power transfer to subsequent low-impedance circuits without significant voltage drop.[17] These traits stem from the signal paths: drain output in CS versus source output in CD, where the latter's feedback mechanism reduces output resistance.[5]
Use cases reflect these strengths, with CS amplifiers employed for core voltage amplification in analog signal chains, such as in operational amplifiers or audio preamps, where high gain is essential.[12] The CD configuration, however, serves primarily as a buffer to match high-impedance sources to low-impedance loads, common in impedance transformation stages like headphone drivers or sensor interfaces.[17] This buffering role prevents loading effects that could degrade prior amplification stages.[5]
Trade-offs between the two underscore their complementary roles: CD excels in impedance transformation and signal isolation, maintaining unity gain to avoid distortion in pass-through applications, but offers no voltage boost and limited current gain beyond what the transistor can supply.[32] Conversely, CS provides superior voltage swing and gain for boosting weak signals, yet its high output impedance may necessitate additional buffering, potentially increasing power consumption or circuit complexity.[12] In high-frequency scenarios, CS can suffer from Miller capacitance effects, while CD's low output impedance aids stability but at the cost of slightly reduced gain below unity due to body effect.[5]
In hybrid multistage designs, a CD stage often follows a CS amplifier to combine their advantages, providing gain from the CS while the CD ensures isolation from load variations and delivers low-impedance output drive.[12] This cascaded approach, sometimes referred to in broader cascode-like contexts for isolation, enhances overall circuit performance by mitigating loading and improving bandwidth without sacrificing the CS's amplification.[32]
With common gate
The common drain (CD) configuration operates as a voltage follower with a voltage gain approximately equal to unity, providing non-inverting amplification where the output at the source closely tracks the input at the gate.[12] In contrast, the common gate (CG) configuration functions primarily as a current buffer with a current gain of approximately -1, meaning it inverts the input current while the voltage gain can be significantly greater than unity, typically A_v \approx g_m (r_o \parallel R_L), where g_m is the transconductance, r_o is the output resistance, and R_L is the load resistance.[12][33]
A fundamental difference lies in their impedance characteristics: the CD amplifier exhibits high input impedance (ideally infinite at the gate) and low output impedance (approximately $1/g_m at the source), making it ideal for isolating high-impedance sources from low-impedance loads.[2] The CG amplifier, however, features low input impedance (approximately $1/g_m at the source) and high output impedance (approximately r_o at the drain), which suits applications requiring current sensing or buffering from low-impedance sources.[12][33]
In terms of use cases, the CD is commonly employed as a source follower for voltage buffering in multistage amplifiers, where it provides impedance matching and signal isolation without significant gain.[12] The CG, on the other hand, is frequently used in cascode input stages to achieve high output impedance and low input impedance, enabling effective current buffering and transimpedance amplification in circuits like low-noise amplifiers.[33][2]
Trade-offs between the two include biasing complexity and frequency response: the CD allows simpler biasing since the source voltage follows the gate, facilitating straightforward DC level setting.[12] The CG requires more careful biasing at the source due to its low input impedance, but it offers superior high-frequency performance owing to a reduced Miller effect—the gate is AC-grounded, minimizing capacitive feedback multiplication compared to other configurations.[2] Both configurations are non-inverting for voltage signals, but the CG uniquely inverts the current signal, which can be advantageous in differential or balanced circuits.[12][33]