Common-base amplifier

The common-base amplifier is the BJT configuration with the input at the emitter, output at the collector, and base at AC ground. It is non-inverting, has a very low input resistance, and — crucially — has no Miller effect, which makes it the high-frequency configuration of choice. It is the BJT analogue of the Common-gate amplifier.

How it works

Hold the base at AC ground and drive the emitter with the input. Looking into the emitter you see the small [[Emitter resistance|$r_e\approx 1/g_m$]] (tens of ohms). An input $v_{in}$ at the emitter therefore drives an emitter current

$i_e=\frac{v_{in}}{r_e}$

Almost all of that becomes collector current, $i_c=\alpha i_e\approx i_e$, which flows through $R_C$ to make the output. Note the sign: pulling the emitter down with $v_{in}$ increases the forward bias of the EBJ, increasing $i_c$, which pulls the collector down too — input and output move the same way. The stage is non-inverting:

$v_c=+i_c{R}_C=\frac{\alpha R_C}{r_e}{v}_{in}\approx g_m{R}_C{v}_{in}\Rightarrow A_v=+g_m{R}_C$

The gain magnitude is the same as the Common-emitter amplifier ($g_m{R}_C$), but with a $+$ sign instead of $-$. (This expression neglects the transistor output resistance $r_o$, taken as $\gg R_C$; $r_o$ is what sets the output resistance discussed below, so the simplification is adopted only for the gain.)

Common-base amplifier: input at emitter, output at collector, base AC-grounded. Non-inverting, gain magnitude similar to CE, low input impedance $r_e$.

Properties and where it is used

• Input resistance: $r_e$ — very low, only tens of ohms. This makes CB poor for a high-impedance voltage source (it heavily loads the source), but excellent as a current buffer: it accepts a current at low impedance and passes it on to a high-impedance collector.
• Output resistance: high (looking into the collector, $\approx r_o$ or higher when degenerated by the source resistance through the transistor).
• No Miller effect. Because the base is grounded, the collector–base capacitance $C_{\mu }$ is not bridged between a swinging input and a swinging output, so it is not multiplied by the gain. This removes the dominant high-frequency pole that limits the CE stage, so the CB amplifier keeps its gain to much higher frequencies. That is the single biggest reason to use it.

These properties make CB the natural stage to put on top of a CE stage in a Cascode amplifier: the CE stage provides the transconductance, the CB stage buffers its output current up to a high impedance without the Miller penalty, giving high gain and wide bandwidth.

The discrete realisation uses the same four-resistor Voltage-divider bias as the CE amplifier, but the signal is coupled into the emitter and the base is held at AC ground by a Bypass capacitor; there is no emitter bypass capacitor because the emitter is the signal node.

Discrete-circuit common-base amplifier — full layout with coupling and bypass capacitors clearly shown.

For the input-at-base, near-unity-gain buffer, see the Emitter follower; for the high-gain inverting workhorse, the Common-emitter amplifier.