I tried to find a good explanation for a source follower but the ones I found were either too vague or too technical, so I'll explain it myself here.
The source follower configuration, also called common drain configuration, is the JFET and MOSFET equivalent of the emitter follower (aka common collector) configuration for bipolar transistors, and the cathode follower (aka common anode) configuration for valves/tubes.
Source follower and emitter follower circuits can be made with active components of either polarity, i.e. N-channel or P-channel FETs, and NPN or PNP transistors. I will describe the N-channel MOSFET version here.
Have a look at the schematic below.
1. It contains one N-channel MOSFET and one resistor (R
S). The input signal is applied to the gate, and the output is taken from the source, which is pulled towards 0V by R
S. The drain is connected to a positive supply rail (VDD), typically around +5 to +12V.
2. The thick line across the bottom of the diagram is the "0V" ("zero volt") rail. It is the reference rail, from which voltages at other points in the circuit (V
G, V
S) are measured. It is also called GND in some circuits, and VSS in JFET- and MOSFET-based circuits. The input and output voltages are also taken relative to this 0V rail.
3. The voltage between the gate and source is marked V
GS. This voltage is important because it is the "input" to the MOSFET: it determines how heavily the MOSFET will conduct through its drain-source path. The MOSFET will start to conduct (through its drain-source path) when V
GS exceeds its gate-source threshold voltage, which is a few volts for a typical MOSFET.
4. If V
GS increases, the MOSFET conducts more heavily; in other words, it allows more current to flow through the drain-source path. If V
GS decreases, the MOSFET conducts less heavily.
5. The gate has an extremely high resistance, so it places almost no load on the driving signal. In other words, almost no current flows into the gate.
That's all you need to know to be able to understand how the source follower works. It's really pretty simple.
Imagine that the input voltage (V
G) is 0V. V
S is pulled to 0V by R
S, and V
GS is also 0V, so the MOSFET does not conduct. Now imagine that the input voltage starts to increase steadily.
When V
G reaches the MOSFET's gate-source threshold voltage, the MOSFET starts to conduct current from VDD into R
S, causing V
S to increase. As V
G increases further, V
GS also increases; this makes the MOSFET conduct more heavily, which causes V
S to increase.
Because of how the MOSFET is connected, it will keep V
GS roughly constant. If the input voltage (V
G) rises, this causes V
GS to increase, which makes the MOSFET conduct more heavily, which pulls V
S upwards, to counter the change in V
GS. If the input voltage (V
G) falls, the dropping V
GS causes the MOSFET to conduct less, so R
S pulls V
S down, keeping V
GS roughly steady.
So the action of the MOSFET tries to keep V
GS constant, and this forces V
S to
follow V
G, with a voltage drop that's roughly equal to the MOSFET's gate-source threshold voltage. That's why the circuit is called a source follower - because the source voltage
follows the gate voltage.
If the source follower is used to buffer an AC signal, the input must have a positive DC bias, to bring the MOSFET into conduction and ensure that the source voltage stays above 0V at all points in the AC signal, even at the maximum signal amplitude. There will be a DC voltage difference between the input and output, but from the point of view of an AC signal, the voltage gain of the circuit is about 1. If you put a 1V RMS sinewave into the input, you get a 1V RMS sinewave at the output. Only the DC level is shifted.
The voltage gain is actually slightly less than 1 because V
GS does vary somewhat as the MOSFET's drain-source current varies.
The source follower does not amplify voltage. Its advantage is that the output current can be a lot higher than the input current. With a JFET or MOSFET, the steady-state input current is almost zero, so a source follower can have a very high current gain. Although it has no voltage gain, it still has power gain because of its current gain, so it is still a very useful circuit.
