How Cascode Amplifiers Eliminate the Miller Effect

This article provides a concise explanation of how a cascode amplifier stage eliminates the Miller effect to significantly extend bandwidth in high-frequency analog electronics. By pairing a common-emitter (or common-source) input stage with a common-base (or common-gate) output stage, the cascode configuration isolates the input from the output. This structural layout drops the voltage gain of the input transistor to nearly unity, preventing the multiplication of parasitic feedback capacitance and preserving high-frequency signal integrity.

Understanding the Miller Effect

In a standard single-transistor amplifier, such as a common-source MOSFET or a common-emitter BJT, there is an inherent parasitic capacitance between the input and output terminals (gate-to-drain or base-to-collector).

According to the Miller theorem, this parasitic capacitance appears at the input of the amplifier multiplied by a factor equal to one plus the inverted voltage gain of the stage (\(C_{Miller} = C_{parasitic} \times (1 + |A_v|)\)). When the amplifier has a high voltage gain, this multiplied capacitance becomes massive. At high frequencies, this large equivalent input capacitance acts as a low-pass filter alongside the input source resistance, drastically reducing the amplifier's bandwidth.

The Cascode Architecture

A cascode amplifier solves this limitation by stacking two transistors in series:

1. The Input Stage: A common-source (FET) or common-emitter (BJT) transistor that receives the input voltage and converts it into a proportional current.
2. The Output Stage: A common-gate (FET) or common-base (BJT) transistor stacked directly on top of the input transistor, which routes the current to the output load resistor.

How the Miller Effect is Eliminated

To eliminate the Miller effect, the voltage gain across the input transistor's parasitic feedback capacitor must be minimized. The cascode configuration achieves this through impedance mismatching between the two stacked transistors.

The output terminal of the bottom (input) transistor is connected directly to the input terminal (source or emitter) of the top transistor. Looking into the source of a common-gate transistor (or emitter of a common-base transistor), the input impedance is extremely low—approximately equal to \(1/g_m\), where \(g_m\) is the transconductance of the top transistor.

Because the bottom transistor drives this very low impedance load (\(1/g_m\)) instead of a large load resistor, the voltage gain of the bottom transistor drops to roughly \(-1\) (\(A_v \approx -g_m \times (1/g_m) = -1\)).

When you apply the Miller equation with a voltage gain of \(-1\): \(C_{Miller} = C_{parasitic} \times (1 + |-1|) = 2C_{parasitic}\)

Instead of the parasitic capacitance being multiplied by a massive voltage gain factor (such as 50 or 100), it is merely doubled. This drastic reduction in total input capacitance pushes the input pole to a much higher frequency.

Achieving High Gain and Wide Bandwidth

Even though the bottom transistor has almost no voltage gain, the overall cascode amplifier still achieves high amplification. The bottom transistor acts as a transconductance amplifier, converting the input voltage into a signal current.

The top transistor acts as a current buffer. It accepts the signal current from the bottom transistor with almost no loss and passes it directly to the high-impedance load resistor at the final output node. Therefore, the high voltage gain takes place exclusively across the top transistor.

Because the gate (or base) of the top transistor is tied to a fixed DC bias voltage (acting as an AC ground), its own parasitic capacitance does not feed back into the input signal path. By separating the transconductance generation from the voltage amplification, the cascode stage successfully delivers both high voltage gain and exceptional high-frequency bandwidth.