A FET amplifier is a circuit that uses a Field Effect Transistor to increase the voltage or current of an input signal by controlling current flow through a semiconductor channel via an electric field rather than an injected base current. In a real circuit, what a FET fundamentally changes is the input impedance; while standard bipolar transistors drag down high-impedance sources, a FET presents an input impedance often exceeding 1MΩ, preserving high-frequency detail from delicate sensors. The most common mistake hobbyists make is confusing FETs with Bipolar Junction Transistors (BJTs). A BJT is a current-controlled device where a small base current dictates a larger collector current, whereas a FET is a voltage-controlled device where the gate-to-source voltage pinches or opens the conductive channel.
The Core Concept: Voltage Control and High Impedance
Think of a FET like a garden hose with a kink in it. You don't need to divert water from the main stream to control the flow; you just apply external physical pressure (the electric field at the gate) to pinch the hose (the semiconductor channel) and restrict the water (the drain current). Because the gate is insulated from the channel by a reverse-biased PN junction in a JFET, or an oxide layer in a MOSFET, virtually zero DC current flows into the gate.
This voltage-control mechanism means we don't use $\beta$ (current gain) to calculate biasing. Instead, we rely on transconductance ($g_m$), which measures how much the drain current changes for a given change in gate voltage, expressed in Siemens (S) or milli-Siemens (mS).
Designing a Common-Source JFET Amplifier: The Math
Let's design a common-source preamplifier using a classic 2N5457 N-channel JFET. We want to amplify a small AC signal from a high-impedance source. Our supply voltage ($V_{DD}$) is 12V, and we want to bias the drain voltage ($V_D$) at roughly half-rail (6V) to allow maximum symmetrical voltage swing.
According to the Electronics Tutorials JFET guide, the DC operating point is governed by Shockley's equation:
I_D = I_{DSS} * (1 - V_{GS} / V_P)^2
The Datasheet Values (Typicals for 2N5457):
- $I_{DSS}$ (Zero gate voltage drain current) = 3.0 mA
- $V_P$ or $V_{GS(off)}$ (Pinch-off voltage) = -2.0 V
Step 1: Choose a target Drain Current ($I_D$)
We'll target $I_D = 1.5$ mA (half of $I_{DSS}$) for a good balance of gain and headroom.
Step 2: Calculate the required Gate-Source Voltage ($V_{GS}$)
Rearranging Shockley's equation:
$V_{GS} = V_P * (1 - \sqrt{I_D / I_{DSS}})$
$V_{GS} = -2.0 * (1 - \sqrt{1.5 / 3.0})$
$V_{GS} = -2.0 * (1 - 0.707) = -0.586$ V
Step 3: Calculate the Source Resistor ($R_S$)
In a self-biased circuit, the gate is held at 0V via a high-value resistor, so the source must sit at +0.586V to create a $V_{GS}$ of -0.586V.
$R_S = V_S / I_D = 0.586V / 1.5mA = 390 \Omega$
Step 4: Calculate the Drain Resistor ($R_D$)
We want $V_D = 6V$. The voltage drop across $R_D$ must be $12V - 6V - 0.586V = 5.414V$.
$R_D = 5.414V / 1.5mA = 3.6 k\Omega$ (Standard E24 value)
Where You Meet FET Amplifiers in Practice
You won't usually find JFETs driving heavy loads; they are the gatekeepers of the signal chain. Here is where specific FET topologies dominate modern bench and field work:
| Application | Preferred FET Type | Common Part Numbers | Why it wins here |
|---|---|---|---|
| RF Front-Ends / Mixers | N-Channel JFET | J310, BF256, MPF102 | Low noise figure and high transition frequency ($f_T$) at VHF/UHF bands. |
| Piezo / Instrument Buffers | N-Channel JFET | 2N5457, J111, TL072 (IC) | Ultra-high input impedance prevents low-pass filtering of high-frequency string harmonics. |
| Class D / Switching Audio | Power MOSFET | IRF510, IRF540N, IRFB4227 | Fast switching times and low $R_{DS(on)}$ minimize dead-time distortion and heat. |
| Oscilloscope Input Stages | Dual JFET / GaAs FET | LSK389, MMBF4416 | Matched pairs provide extremely low input bias current (<10pA) for accurate high-Z probing. |
Bench Walkthrough: Building a Piezo Buffer (And How It Fails)
Let's take the 2N5457 math from above and apply it to a real-world scenario: building a unity-gain source-follower buffer for an acoustic guitar's piezo under-saddle pickup. Piezo disks have a source impedance of roughly 1MΩ to 3MΩ. If you plug one directly into a standard 10kΩ BJT amplifier input, the RC low-pass filter formed by the piezo's internal capacitance and the amp's input resistance will roll off everything above 2kHz, making the guitar sound like it's wrapped in a blanket.
- Setup: I wired a 2N5457 in a common-drain (source follower) configuration on a breadboard. I used a 2.2MΩ gate pulldown resistor to set the input impedance, a 10µF coupling capacitor on the input, and my calculated 390Ω source resistor tied to ground. The drain was tied directly to a fresh 12V bench supply.
- Numbers: According to the math, the source voltage should sit at 0.586V, and the output should track the gate voltage with a gain of ~0.95.
- Outcome: I clipped the piezo to the bridge, plugged the output into my audio interface, and struck a chord. The output was a flat, heavily distorted square wave with massive 60Hz hum.
- What Went Wrong: I fell victim to JFET parameter spread. The All About Circuits semiconductor textbook explicitly warns about this: the 2N5457 datasheet lists $I_{DSS}$ anywhere from 1.0mA to 5.0mA. The specific transistor I grabbed from my bin had an $I_{DSS}$ of 4.8mA. My 390Ω resistor wasn't nearly large enough to pinch the channel down to a safe operating point. The JFET biased at roughly 3.8mA, pulling the source voltage up to 1.5V and driving the transistor deep into the ohmic (linear) region. It was acting as a resistor, not an amplifier, clipping the negative half of the audio waveform entirely.
Frequently Asked Questions
Do I need a gate resistor if the gate draws zero current?
Yes, absolutely. While the DC gate current is essentially zero, the gate still needs a DC path to ground to establish the 0V reference for the $V_{GS}$ bias. Without a high-value gate resistor (typically 1MΩ to 10MΩ), the gate is floating. Stray electrostatic charges will accumulate on the gate capacitance, slowly shifting the bias voltage until the FET either pinches off entirely or saturates, killing your signal.
Why use a JFET instead of a MOSFET for small audio signals?
MOSFETs have an insulated gate, which gives them even higher input impedance, but they suffer from higher gate capacitance and a more abrupt, harsh clipping characteristic when overdriven. JFETs have a softer, more gradual clipping curve that mimics vacuum tubes, which is why they are heavily favored in guitar overdrive pedals and high-end audio preamps. Furthermore, JFETs generally exhibit lower 1/f (flicker) noise at audio frequencies compared to standard small-signal MOSFETs.
What is the purpose of a source bypass capacitor?
If you place a capacitor in parallel with your source resistor ($R_S$), it acts as a short circuit for AC signals while leaving the DC bias untouched. This prevents 'source degeneration' (local negative feedback). Without the bypass capacitor, your AC signal has to push through $R_S$, which reduces the overall voltage gain of the amplifier. With the capacitor, the AC gain increases significantly, approaching $g_m * R_D$.
