Unity-gain phase splitter

In certain applications, it’s beneficial to produce a signal and its exact opposite—two signals that are 180 degrees out of phase. This can be accomplished using an emitter-degenerated amplifier configured for a gain of −1, as illustrated in circuit diagram below.

To ensure maximum symmetrical output swing without clipping at either output, the quiescent collector voltage is set to 0.75VCC rather than the more common 0.5VCC. This setup allows:

• The collector to swing between 0.5VCC and VCC.

• The emitter to swing between ground and 0.5VCC.

A critical consideration is that the phase-splitter outputs must drive equal impedances (or very high impedances) at both outputs. If the loads are mismatched, the symmetry between the two signals will be compromised, affecting the phase relationship and gain balance.

Below shows the waveform of the input signal and the phase shifted signal at the collector and emitter.

 

A clever application of the phase splitter is demonstrated in the circuit diagram below. 

 This circuit generates an output signal with an adjustable phase shift (ranging from 0° to 180°) while maintaining a constant amplitude. It is basically a signal phase shifting circuit. The behavior of the circuit can be best understood using a phasor diagram of voltages. If the input signal is represented as a unit vector along the real axis, the resulting signals appear as shown in figure below.

In the phasor diagram, the signal vectors $v_R$ (across the resistor) and $v_C$ (across the capacitor) must always be perpendicular to each other. These vectors combine to form a resultant vector of constant length along the real axis. A geometric theorem states that the locus of such points forms a circle. This ensures that the resultant vector (the output voltage) always has a unit length, meaning the output amplitude remains equal to the input amplitude. At the same time, the phase of the output can vary from nearly 0° to 180° relative to the input signal as the potentiometer $R$ is adjusted from nearly zero to a value much larger than the capacitive reactance $X_C$ at the operating frequency.

For example, consider a circuit operating at $f=1\text{kHz}$ with $C=0.1\mu \text{F}$. To achieve a phase shift approaching $180^{\circ }$, the resistance $R$ must be sufficiently large. Using the relationship $\theta =2{\tan }^{-1}(wRC)$, where $w=2\pi f$, we calculate $R\approx 1.59\text{M}\mathrm{\Omega }$ for $\theta \approx 180^{\circ }$. To allow for adjustable phase shifts, a 2 MΩ potentiometer is suitable, as it provides the necessary range to vary $R$ and achieve the desired phase shift.

However, it’s important to note that the phase shift depends on the frequency of the input signal for a given setting of the potentiometer $R$. This frequency dependence is a key consideration when designing such circuits. While a simple RC highpass or lowpass network could also be used as an adjustable phase shifter, its output amplitude would vary significantly as the phase shift is adjusted, making it less practical for applications requiring constant amplitude.

Another consideration is the loading effect on the phase-splitter circuit. Ideally, the load impedance should be much larger than the collector and emitter resistors to avoid disrupting the circuit’s operation. As a result, this circuit has limited utility in applications requiring a wide range of phase shifts, particularly when driving low-impedance loads. For improved phase splitter see balanced phase splitter with op-amp. If you like to build phase splitter consider using the online phase shifter calculator.