common source amplifier
A common-source amplifier is the MOSFET's direct analogue to the bipolar common-emitter stage: the source terminal is common to input and output, the signal drives the gate, the output comes off the drain through a load resistor or current-source load, and the output is an inverted, amplified copy of the input — but because the gate draws essentially zero DC current rather than a beta-dependent base current, the common-source stage sidesteps the single biggest headache of its bipolar counterpart and became the default gain stage once CMOS processes made MOSFETs cheap and uniform enough to dominate analog design.
Transconductance gm, not a current-gain factor like beta, is what sets the gain — and that single difference changes how the whole stage is designed. A bipolar common-emitter stage's gain depends on re, which is tied to emitter current and the transistor's (somewhat variable) characteristics; a common-source stage's gain depends on gm, which for a MOSFET in saturation is proportional to the square root of the bias current times the process transconductance parameter, giving gm = 2 x ID / (VGS - VT) — a relationship that is more predictable across a production run precisely because MOSFET threshold voltage and transconductance parameters are controlled directly by the fabrication process rather than by a current-gain factor that can vary two-to-one between individual bipolar parts.
Near-infinite DC input impedance is the property that made CMOS logic itself possible, and it falls straight out of the common-source topology. Because the gate is insulated from the channel by a thin oxide layer, no DC current flows into it at all (aside from negligible leakage), so a common-source stage places essentially zero steady-state load on whatever drove its gate — this is exactly why CMOS logic gates can be cascaded by the millions on a single die without needing bias networks to supply base current at every stage, something a bipolar common-emitter chain could never do at that density.
The Miller effect still applies, and it is often worse, not better, than in a bipolar stage. The gate-to-drain capacitance Cgd gets multiplied by the stage's gain just like a bipolar transistor's base-collector capacitance does, and because MOSFETs built for high transconductance often have larger gate capacitance than an equivalently-sized bipolar transistor's junction capacitance, common-source stages frequently need deliberate techniques — cascode stages, smaller device sizing traded against gain, or explicit Miller compensation capacitors — to keep bandwidth where a bipolar designer might get away with ignoring the issue.
It is the universal analog building block inside modern ICs precisely because the whole chip is already built from MOSFETs. Operational amplifiers, ADC and DAC input stages, LNA front ends, and the vast majority of on-chip analog gain in any CMOS process use common-source stages as their basic amplifying element, often stacked into cascode or folded-cascode structures to push gain, bandwidth, or output swing further — the same way common-emitter is the default bipolar building block, common-source is the default CMOS one, which is why understanding this one stage unlocks reading the vast majority of analog IC schematics.
Read a common-source amplifier through a *zero-DC-input-loading* lens rather than just a *MOSFET version of common-emitter* lens: swapping a current-controlled base for a voltage-controlled, current-isolated gate is what let gain stages be stacked into the enormous, densely cascaded analog and mixed-signal blocks that fill a modern chip, something the bipolar common-emitter stage's base-current demand could never have scaled to in the same way.