Self bias vs fixed bias in MOSFET

In fixed (gate-bias) biasing the gate voltage is set directly (for example by a fixed supply or a divider with no source resistor), so VGS is constant. In self (source-resistor) biasing the gate is held by a divider but a resistor RS in the source makes VGS = VG − ID RS, which feeds the drain current back to the input.

Why self-bias is better

1. Negative feedback stabilises ID. If ID rises (temperature, device change), the drop ID RS rises, VGS falls, and ID is pulled back. In fixed bias there is no such correction.
2. Less sensitivity to device spread. The threshold voltage Vt and constant k vary widely between devices of the same type. Fixed bias lets ID = k (VGS − Vt)² change a lot; self-bias holds it near the design value.
3. Stable Q point and gain. The Q point stays in saturation, so distortion and gain variation are lower. With a bypass capacitor across RS, ac gain is not reduced.
4. Needs only one supply in the divider form.

Numerical illustration

MOSFET with k = 0.5 mA/V² and Vt = 1 V, designed for ID = 0.5 mA (VGS = 2 V). Suppose Vt changes to 1.5 V.

Self-bias therefore holds the Q point far better. The cost is a lower dc voltage left for the drain-source (VDS reduces by ID RS).