A common source amp is a field-effect transistor (FET) circuit configuration where the source terminal is grounded (or common to both input and output), providing high voltage gain and a 180-degree phase inversion. In a real circuit, this topology changes a high-impedance, low-voltage signal (like a piezo sensor or guitar pickup) into a lower-impedance, higher-voltage swing capable of driving subsequent stages, while simultaneously flipping the signal's phase. It is the FET equivalent of the BJT common emitter amplifier, and it is the workhorse of voltage amplification in modern electronics.
The Core Concept: What a Common Source Amp Actually Does
When you apply a small AC voltage to the gate of an N-channel MOSFET, it modulates the channel's conductivity. Because the drain is tied to a positive supply through a load resistor ($R_D$), any increase in gate voltage causes an increase in drain current. This higher current drops more voltage across $R_D$, which forces the voltage at the drain node to decrease. Conversely, a lower gate voltage reduces drain current, allowing the drain voltage to rise. This inverse relationship is what creates the 180-degree phase shift (signal inversion) and the voltage amplification.
People commonly confuse the common source amp with the common drain (or source follower) configuration. In a common drain setup, the output is taken from the source terminal, resulting in a voltage gain of slightly less than 1 (unity gain) and zero phase inversion. The common drain is used for impedance buffering, whereas the common source amp is strictly used when you need to multiply the voltage amplitude of your signal.
The Math That Matters: A Worked Numeric Example
Let's calculate the mid-band voltage gain ($A_v$) of a common source amp using a widely available 2N7000 N-channel MOSFET. We will ignore the Early effect and assume the output resistance of the MOSFET ($r_o$) is much larger than our drain resistor.
- Establish the DC Bias Point: We set our supply voltage ($V_{DD}$) to 12V. Through our gate bias network, we bias the gate to yield a quiescent drain current ($I_D$) of 5mA.
- Set the Drain Resistor ($R_D$): We choose a standard 1kΩ resistor for $R_D$. The DC voltage drop across $R_D$ is $V = I \times R = 5mA \times 1000\Omega = 5V$. This leaves the quiescent drain voltage ($V_{DS}$) at $12V - 5V = 7V$, giving us plenty of headroom for the AC swing.
- Determine Transconductance ($g_m$): Transconductance dictates how much the drain current changes for a given change in gate voltage. For a 2N7000 biased at 5mA in the saturation region, the small-signal $g_m$ is approximately 50 mS (millisiemens), or 0.05 A/V.
- Calculate Voltage Gain: Multiply $g_m$ by $R_D$.
$A_v = -0.05 \text{ A/V} \times 1000\Omega = $ -50 V/V.
The negative sign confirms the 180-degree phase inversion. If you feed a 10mV peak-to-peak sine wave into the gate, you will get a 500mV peak-to-peak sine wave at the drain, flipped upside down.
Where You Meet This in Practice
You will rarely build a discrete 2N7000 common source amp for high-fidelity audio today, as op-amps handle that job with fewer components. However, this topology is everywhere in specific niches:
- RF Front Ends: Low Noise Amplifiers (LNAs) in Wi-Fi routers and SDR (Software Defined Radio) dongles use common source configurations (often in cascode arrangements) to amplify microvolt antenna signals without adding excessive thermal noise.
- Guitar Effects Pedals: 'MOSFET booster' pedals (like the classic AMZ Mosfet Booster) use a single common source stage to provide a gritty, tube-like voltage gain before hitting a clipping stage.
- Piezo Sensor Buffers: Piezo vibration sensors have massive output impedance. A common source amp (sometimes modified with high-value gate resistors) provides the necessary high input impedance to prevent loading the sensor down, while stepping up the voltage for an ADC.
Bench Walkthrough: When Biasing Goes Wrong
Theory assumes perfect biasing. On the bench, ignoring the threshold voltage ($V_{GS(th)}$) of your specific MOSFET will ruin your signal. Here is a real-world scenario from the workbench.
Setup: Building a simple electret microphone preamp using a BS170 MOSFET. The circuit uses a 9V battery, a 4.7kΩ drain resistor ($R_D$), and a 1kΩ source resistor ($R_S$) for DC stability, bypassed with a 10µF capacitor. The gate is biased via a voltage divider.
The Numbers: I aimed for a quiescent drain current ($I_D$) of 1mA. With $R_D = 4.7k\Omega$, the voltage drop across the drain resistor is 4.7V. The voltage drop across $R_S$ is 1V. Therefore, the quiescent $V_{DS}$ (drain-to-source voltage) should be $9V - 4.7V - 1V = 3.3V$. The theoretical AC gain is roughly $-R_D / (1/g_m) \approx -20$.
The Outcome: When speaking into the mic, the oscilloscope showed the negative half of the AC waveform at the drain clipping flat against the 9V rail, while the positive half looked clean. The audio output sounded heavily distorted and 'farty'.
What Went Wrong: The BS170 has a relatively high and widely varying $V_{GS(th)}$ (typically 2.5V, but can range from 2.0V to 3.5V). My voltage divider was designed assuming a $V_{GS}$ of 2.5V. The specific transistor I grabbed from the bin had a $V_{GS(th)}$ of 3.2V. Because the gate voltage was too low to fully open the channel, the actual $I_D$ was only 0.4mA, not 1mA.
This shifted the DC bias point. The actual voltage drop across $R_D$ was only 1.88V, pushing the quiescent drain voltage up to nearly 6.1V. When the AC signal tried to swing positive (which means the drain voltage swings negative towards ground), it had plenty of room. But when the AC signal swung negative (drain voltage swings positive towards $V_{DD}$), it immediately hit the 9V ceiling and clipped.
The Fix: I replaced the fixed gate voltage divider with a 1MΩ potentiometer, adjusted it while monitoring the drain voltage with a multimeter until $V_D$ sat exactly at $4.5V$ (half of $V_{DD}$), and the clipping vanished. Always design for the specific $V_{GS}$ of the physical part in your hand, or use source-degeneration feedback to stabilize the bias point against part-to-part variations.
Common Source vs. Common Drain: Clearing Up the Confusion
If you are debugging a circuit and the signal isn't amplifying, you might have wired a common drain by mistake. Here is how to tell them apart at a glance.
| Feature | Common Source Amp | Common Drain (Source Follower) |
|---|---|---|
| Output Node | Drain | Source |
| Voltage Gain ($A_v$) | High (e.g., -20 to -100) | Unity (slightly less than 1) |
| Phase Shift | 180° (Inverted) | 0° (Non-inverted) |
| Input Impedance | Very High | Very High |
| Output Impedance | Moderate to High (depends on $R_D$) | Very Low (excellent for driving loads) |
| Primary Use Case | Voltage amplification | Current buffering / Impedance matching |
Frequently Asked Questions
Why does my common source amp oscillate at high frequencies?
This is usually caused by the Miller Effect. The parasitic capacitance between the gate and drain ($C_{gd}$) is multiplied by the voltage gain, creating a massive effective input capacitance that can interact with your source impedance to cause phase shifts and high-frequency instability. To fix this, keep gate wiring short, use a small gate-stopper resistor (e.g., 100Ω) directly at the gate pin, or add a small compensation capacitor across the drain-to-gate to roll off high-frequency gain.
Do I need a source bypass capacitor?
If you want maximum AC voltage gain, yes. A resistor at the source ($R_S$) is great for stabilizing the DC bias point against temperature and part variations, but it introduces negative feedback for AC signals, which crushes your gain. Placing a large electrolytic capacitor (e.g., 10µF to 100µF) in parallel with $R_S$ shorts the AC signal to ground, restoring your full $g_m \times R_D$ gain while keeping the DC stability intact.
Can I use a JFET instead of a MOSFET?
Yes. A JFET (like the J201 or 2N5457) wired in a common source configuration operates on the exact same topological principles. The main difference is that JFETs are normally-on (depletion-mode) devices, meaning they require a negative gate-to-source voltage to pinch off the channel and set the bias point, whereas enhancement-mode MOSFETs like the 2N7000 require a positive gate voltage to turn on. For further reading on FET topologies, refer to this comprehensive guide on MOSFET amplifiers or the common source topology overview on Wikipedia.
