Capacitively Loaded Emitter Follower PUBLIC

An emitter follower is supposed to have its output voltage follow changes in its input approximately 1 to 1, but fast input edges and a capacitively loaded output lead to undesired and asymmetric behavior.

V1 provides a static 2.5V offset to bias the emitter follower near the center of its range.

V2 is a piecewise-linear source which generates a repeating signal with a positive-going +1V pulse and negative-going -1V pulse. All transitions have slope 1V / 0.1ms, or 10000 V/s.

Run the time-domain simulation and see how different the output responds to the positive-going and negative-going edges! This is a case where the small-signal frequency-domain analysis doesn't tell the whole picture, because the input level jumps are big enough and fast enough to enter a dramatically different regime of transistor operation. That is, when V(in) drops suddenly, Q1 shuts off, so the charge on capacitor C1 can only discharge slowly through R1 (a relatively high resistance path). But when V(in) rises suddenly, Q1 turns on harder, so C1 can quickly charge. This leads to the asymmetrical behavior in the time-domain simulation.

A few things to try in CircuitLab

DC

The DC simulation shows operating point bias conditions, including power dissipation in the transistor.

DC Sweep

The DC sweep is set up to sweep over the transistor's forward beta, or current gain. This shows how the bias point of the circuit is not very sensitive to the transistor's beta -- the current I(Q1.nC) and V(out) stay about the same regardless of betas between 100 and 200.

Time domain

The time-domain simulation shows the input voltage and output voltage, and the inability of the emitter follower to follow the negative-going pulses is clear. A third plot shows something like the "error" signal, "V(in) - V(out) - 0.6", using CircuitLab's Advanced Graphing mode to display the voltage follower error on a separate plot.

Frequency domain

The frequency-domain analysis shows the small-signal transfer function for various capacitive load values. However, this is clearly a circuit where the small-signal behavior doesn't tell the whole story!