MOSFET biasing and small-signal model

MOSFET regions, square law, divider-plus-source-resistor biasing, and the small-signal model with g_m, r_o and common-source gain.

Drafted with Aria, reviewed by the AiCanCode.org team. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

MOSFETs are the transistors inside almost every op-amp, ADC front end and microcontroller you will use in instrumentation, and discrete MOSFETs drive heaters, valves and motors. Their near-infinite DC gate resistance makes them ideal input devices for high-impedance sensors such as pH electrodes and piezo transducers. As with the BJT, you first fix a stable DC operating point and then replace the device by a linear small-signal model.

Key ideas

Structure and threshold. In an n-channel enhancement MOSFET no channel exists at V_GS = 0. When V_GS exceeds the threshold voltage V_t, an inversion layer of electrons forms under the gate and links source and drain. The gate is insulated by oxide, so the DC gate current is essentially zero. Depletion-type MOSFETs have a built-in channel (negative V_t for n-channel); p-channel devices work the same way with all polarities reversed.

Regions of operation (n-channel). Define the overdrive V_ov = V_GS − V_t.

  • Cut-off: V_GS < V_t, I_D ≈ 0.
  • Triode (linear): V_GS > V_t and V_DS < V_ov. The channel behaves like a voltage-controlled resistor; used for switches and analogue multiplexers.
  • Saturation (pinch-off): V_GS > V_t and V_DS ≥ V_ov. I_D depends on V_GS but only weakly on V_DS, so the device is a voltage-controlled current source. This is the amplifying region. (Note the naming clash: MOSFET "saturation" corresponds to the BJT active region.)

Square law. In saturation I_D = ½·μ_n·C_ox·(W/L)·V_ov². Many textbooks write k_n = ½·μ_n·C_ox·(W/L) so that I_D = k_n·V_ov². Read each question carefully to see which convention its "k" uses.

Channel-length modulation. As V_DS rises the pinch-off point moves towards the source, shortening the effective channel, so I_D rises slightly. This is modelled by the factor (1 + λ·V_DS) and gives a finite output resistance r_o = 1/(λ·I_D).

Biasing. Because V_t and k_n vary between devices and with temperature, fixing V_GS directly gives a poorly controlled I_D. The standard discrete scheme sets the gate voltage with a divider (no gate current, so the divider is exact) and adds a source resistor R_S. The equation V_GS = V_G − I_D·R_S, combined with the square law, gives a quadratic; keep the root with V_GS > V_t. R_S provides negative feedback: if I_D rises, V_GS falls. Always check V_DS ≥ V_ov afterwards.

Small-signal model. Gate open circuit (r_in = ∞ at low frequency), current source g_m·v_gs from drain to source, and r_o in parallel. Unlike the BJT there is no r_π. At high frequency C_gs and C_gd are added. If the body is not tied to the source, a second source g_mb·v_bs models the body effect.

Temperature. Mobility falls and V_t falls as temperature rises. At normal bias currents the mobility effect wins, so I_D falls with temperature, which is why MOSFETs do not suffer thermal runaway and can be paralleled more easily than BJTs.

Formulas

I_D = k_n·(V_GS − V_t)²·(1 + λ·V_DS) with k_n = ½·μ_n·C_ox·(W/L) — saturation; k_n in A/V², λ in V⁻¹.

I_D = 2k_n·[(V_GS − V_t)·V_DS − V_DS²/2] — triode region.

V_DS ≥ V_GS − V_t — condition for saturation.

g_m = 2·I_D / V_ov = 2·k_n·V_ov = 2·√(k_n·I_D) — transconductance (S).

r_o = 1 / (λ·I_D) — output resistance (Ω).

A_v = −g_m·(R_D ∥ R_L ∥ r_o) — common-source gain, source bypassed.

A_v = −g_m·R_D / (1 + g_m·R_S) — common-source gain with unbypassed R_S (r_o neglected).

Worked examples

Example 1 (standard: bias). An n-channel enhancement MOSFET has V_t = 1 V and k_n = 1 mA/V² (I_D = k_n·V_ov²). A divider sets V_G = 4 V, R_S = 1 kΩ, R_D = 3 kΩ, V_DD = 12 V. Find I_D, V_GS and V_DS.

  1. Gate current is zero, so V_GS = V_G − I_D·R_S = 4 − I_D (I_D in mA, R_S in kΩ).
  2. Saturation: I_D = k_n·(V_GS − 1)². Let x = V_GS − 1; then I_D = x² and x = 3 − x².
  3. Solve x² + x − 3 = 0: x = (−1 + √13)/2 = 1.303 V (the negative root gives V_GS < V_t and is rejected).
  4. I_D = 1.303² = 1.697 mA; V_GS = 2.303 V.
  5. V_DS = V_DD − I_D·(R_D + R_S) = 12 − 1.697 × 4 = 5.21 V. Since 5.21 V > V_ov = 1.30 V, the device is in saturation.

Answer: I_D ≈ 1.70 mA, V_GS ≈ 2.30 V, V_DS ≈ 5.21 V.

Example 2 (GATE level: small signal). For Example 1 take λ = 0.02 V⁻¹ (ignore its effect on bias). Find g_m, r_o and the voltage gain with no load, (a) with R_S bypassed and (b) with R_S unbypassed.

  1. g_m = 2I_D/V_ov = 2 × 1.697 mA / 1.303 V = 2.61 mS.
  2. r_o = 1/(λI_D) = 1/(0.02 × 1.697×10⁻³) = 29.5 kΩ.
  3. (a) A_v = −g_m(R_D ∥ r_o): 3 kΩ ∥ 29.5 kΩ = 2.72 kΩ, so A_v = −2.61 mS × 2.72 kΩ = −7.09.
  4. (b) A_v = −g_m·R_D/(1 + g_m·R_S) = −(2.61 × 3)/(1 + 2.61 × 1) = −7.82/3.61 = −2.17.

Answer: g_m ≈ 2.6 mS, r_o ≈ 29.5 kΩ, A_v ≈ −7.1 (bypassed) and ≈ −2.2 (unbypassed). MOSFET stages give less gain than BJT stages at the same current, because g_m = 2I_D/V_ov is much smaller than I_C/V_T.

Common mistakes

  • Mixing the two conventions for k (with and without the ½) and getting I_D wrong by a factor of 2.
  • Keeping the wrong root of the bias quadratic; always reject the root with V_GS < V_t.
  • Forgetting to check V_DS ≥ V_GS − V_t after solving.
  • Calling the MOSFET saturation region "saturation" in the BJT sense (switch fully ON); for a MOSFET the fully-ON switch is the triode region.
  • Using g_m = I_D/V_T (BJT formula) for a MOSFET in strong inversion.
  • Ignoring the body effect when the source is not tied to the body (for example, in a source follower in an IC).

For GATE IN

Expect region identification from terminal voltages, bias-point calculations that lead to a quadratic, g_m and r_o from given k_n and λ, and common-source gain with and without source degeneration. Also expect conceptual questions on enhancement versus depletion devices, channel-length modulation and why MOSFET current falls with temperature. Practise solving the bias quadratic quickly and checking the region.

Quick check

  1. V_t = 0.5 V, V_GS = 1.5 V, V_DS = 0.6 V. Which region?
  2. I_D = 1 mA at V_ov = 0.25 V. What is g_m?
  3. With λ = 0.01 V⁻¹ and I_D = 2 mA, what is r_o?
  4. Why is a MOSFET's DC input resistance so high?

Answers: 1. Triode (V_DS < V_ov = 1 V). 2. 8 mS. 3. 50 kΩ. 4. The gate is insulated from the channel by oxide, so no DC gate current flows.

Try answering each one aloud before you open it.

  1. 1.What is MOSFET biasing and why is it important in analog electronics?Concept

    MOSFET biasing refers to setting the DC operating voltage and current of a MOSFET to ensure it operates in the desired region, typically the saturation region for analog applications. This is important because it stabilizes the MOSFET's operation, allowing it to amplify signals without distortion. Proper biasing ensures that the MOSFET remains in the active region over the entire input signal range.

  2. 2.Explain the small-signal model of a MOSFET.Concept

    Around a saturation-region bias point the MOSFET is linearised as an open-circuit gate (infinite input resistance at low frequency), a voltage-controlled current source g_m·v_gs from drain to source, and an output resistance r_o = 1/(λ·I_D) in parallel. The transconductance is g_m = 2I_D/V_ov = 2√(k_n·I_D). At high frequency the gate capacitances C_gs and C_gd are added, and if the body is not tied to the source a second source g_mb·v_bs models the body effect.

  3. 3.How does the threshold voltage affect MOSFET biasing?Concept

    The threshold voltage (Vth) is the minimum gate-to-source voltage required to create a conducting path between the source and drain. It affects MOSFET biasing because it determines the gate voltage needed to turn the MOSFET on. A higher threshold voltage requires a higher gate voltage for the same level of conduction, impacting the design of biasing circuits.

  4. 4.Why is the saturation region preferred for MOSFET operation in analog circuits?Application

    The saturation region is preferred for MOSFET operation in analog circuits because it provides a constant current source characteristic, which is ideal for amplification. In this region, the MOSFET's drain current is relatively independent of the drain-source voltage, allowing for linear amplification of the input signal. This stability is crucial for maintaining signal integrity.

  5. 5.What happens if a MOSFET is improperly biased in an analog circuit?Application

    If a MOSFET is improperly biased, it may not operate in the desired region, leading to distortion or clipping of the output signal. For example, if biased in the cutoff region, the MOSFET will not conduct, resulting in no amplification. If biased in the triode region, the MOSFET may not provide the necessary gain, affecting the circuit's performance.

  6. 6.How does temperature affect MOSFET biasing?Application

    Two effects compete: the threshold voltage falls with temperature (which would raise I_D) and the carrier mobility falls (which lowers I_D). At normal operating currents the mobility effect dominates, so drain current decreases as the device heats up. This negative temperature coefficient means MOSFETs do not suffer thermal runaway and share current well in parallel, though the Q-point still drifts and a source resistor is used to stabilise it.

  7. 7.Why is a resistor often used in the source terminal of a MOSFET in biasing circuits?Application

    A resistor in the source terminal, known as source degeneration, is used to stabilize the biasing point against variations in threshold voltage and transconductance. It provides negative feedback, which helps maintain a constant current through the MOSFET despite changes in temperature or device parameters, improving the linearity and stability of the amplifier.

  8. 8.Calculate the transconductance (gm) of a MOSFET with a drain current (Id) of 2 mA and an overdrive voltage (Vov) of 0.2 V.Numerical

    The transconductance (gm) can be calculated using the formula gm = 2 * Id / Vov. Substituting the given values: gm = 2 * 2 mA / 0.2 V = 20 mS (millisiemens).

  9. 9.A MOSFET has a threshold voltage (Vth) of 1 V and is biased with a gate-source voltage (Vgs) of 3 V. Calculate the overdrive voltage (Vov).Numerical

    The overdrive voltage (Vov) is calculated as Vov = Vgs - Vth. Substituting the given values: Vov = 3 V - 1 V = 2 V.

  10. 10.Explain how the body effect influences MOSFET biasing.Application

    The body effect refers to the change in the threshold voltage of a MOSFET due to a voltage difference between the body (substrate) and the source terminal. This effect can increase the threshold voltage, requiring a higher gate-source voltage to maintain the same level of conduction. It is important to consider in biasing circuits, especially in integrated circuits where the body is often connected to a fixed potential.

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?