Lock-in amplifiers are indispensable instruments for extracting and measuring weak signals buried in high levels of noise. These devices are widely used in disciplines such as physics, materials science, and biomedical engineering. Traditional commercial lock-in amplifiers offer exceptional performance but come with significant complexity and cost. With the availability of high-performance operational amplifiers (op amps), it is now possible to design and build active lock-in amplifiers that provide sensitive signal detection at a fraction of the price. This article explores the fundamental principles of lock-in amplification, the role of op amps, detailed circuit design stages, practical building steps, and key considerations for achieving high sensitivity.

The Principle of Lock-In Amplification

Lock-in amplification relies on phase-sensitive detection (PSD). The technique multiplies the incoming signal, which may contain noise, with a reference waveform that is at the same frequency as the desired signal. The multiplication produces a product containing a DC component proportional to the amplitude of the signal and the cosine of the phase difference between the signal and the reference. A low-pass filter then removes all AC components, including the sum frequency and any noise away from the reference frequency. This process allows only the coherent component at the reference frequency to pass, effectively rejecting broadband noise. The mathematical foundation is simple: if Vin(t) = A sin(ωt + φ) and the reference is Vref(t) = sin(ωt), the product is (A/2)[cos(φ) - cos(2ωt + φ)]. After low-pass filtering, the output is (A/2) cos(φ), which carries the amplitude information.

Why Use Operational Amplifiers?

Operational amplifiers offer a versatile platform for implementing each stage of a lock-in amplifier. Op amps provide high open-loop gain, low input bias currents, and excellent common-mode rejection, making them ideal for precision analog processing. They can be configured as multipliers, active filters, and voltage amplifiers within a single compact circuit. Modern op amps such as the OPA2277 or the low-noise OPA1612 enable designs with very low input noise, which is critical for detecting signals far below the noise floor. The use of op amps also simplifies parameter adjustments — resistor and capacitor values can tune cutoff frequencies, gain, and phase response. Compared to discrete transistor circuits, op amp designs are more predictable, easier to debug, and readily adaptable to different frequency ranges.

Core Stages of an Active Lock-In Amplifier

1. Mixing Stage

The mixer performs the multiplication of the input signal with the reference. For high linearity and wide bandwidth, an analog multiplier IC like the AD633 (Analog Devices) is commonly used. This four-quadrant multiplier outputs the product of its two inputs with an internal scale factor. The op amps within the AD633 can be supplemented with external op amps for buffer and offset adjustment. Alternatively, a switching mixer can be built using an op amp as a comparator driving a precision analog switch while a second op amp provides differential output. This topology offers low offset and good dynamic range but may introduce switching noise. For very low frequency applications (< 1 kHz), a simple op amp multiplier circuit using a Gilbert cell configuration can be constructed, though it requires careful matching of components. The mixer output must be well balanced to minimize feedthrough of the reference signal into the signal path.

2. Filtering Stage

Following the mixer, a low-pass filter (LPF) removes the undesired sum frequency components and out-of-band noise. The filter’s cutoff frequency determines the measurement bandwidth and thus the time constant of the lock-in amplifier. A second-order active filter, such as the Sallen-Key topology, provides a sharp roll-off with minimal component count. Choosing a low-noise op amp for the filter stage is essential to prevent additional noise contributions. For a time constant of τ, a cutoff frequency fc = 1/(2πτ) is typical. A four-pole filter can be constructed by cascading two second-order sections to achieve steeper attenuation, which improves noise rejection. The design trade-off includes the filter settling time versus noise bandwidth; slower filters yield higher noise rejection but limit the measurement speed.

3. Amplification and Output

The filtered DC signal is generally very small and requires further amplification before it can be measured by an ADC or meter. A non-inverting op amp stage provides precise gain determined by resistor ratios. For differential output or to cancel any residual offset, an instrumentation amplifier such as the INA128 can be employed. The output stage should be buffered to drive subsequent stages without loading the filter. Offset nulling can be achieved by injecting a small DC voltage through a summing circuit. The final output is usually proportional to the amplitude of the measured signal times the cosine of the phase difference between the signal and reference. If the phase is unknown, a second channel using a quadrature reference (90-degree shifted) allows the independent extraction of amplitude (R = √(X² + Y²)) and phase.

Design Considerations for High Sensitivity

Noise Sources

The overall noise performance of the lock-in amplifier is determined by the noise voltages and currents of the op amps, the thermal noise of resistors, and power supply ripple. Selecting op amps with low voltage noise density (e.g., < 1 nV/√Hz) and low current noise is critical for the first stages. Resistor values should be kept low (typically below 10 kΩ) to minimize Johnson noise. Shielding the entire circuit in a grounded metal enclosure and using decoupling capacitors (0.1 µF ceramic in parallel with 10 µF electrolytic) at each op amp power pin reduces external interference. A star ground topology, where all ground connections meet at a single point, helps prevent ground loops that could introduce low-frequency noise.

Dynamic Reserve and Full Scale

Dynamic reserve refers to the ability of a lock-in amplifier to handle a large noise background without saturating the signal chain. In an op amp-based design, the dynamic reserve is limited by the available headroom of the power supply and the linear range of the multiplier. Pre-amplification with a bandpass filter centered on the reference frequency can increase the reserve by attenuating noise before the mixer. The full-scale input range should be chosen such that the maximum expected signal plus noise does not exceed the linear operating region of any stage.

Phase Adjustment

For accurate amplitude measurement when the signal phase is unknown or varies, a dual-phase lock-in amplifier is necessary. This is accomplished by generating both a sine and cosine reference waveform (e.g., using a quadrature oscillator or a DDS with two outputs). Two separate mixer and filter channels process the input, yielding an in-phase (X) and quadrature (Y) output. The amplitude is then calculated as √(X² + Y²). This approach eliminates the need for manual phase adjustment and provides full phase-insensitive detection.

Building a Practical Circuit

To construct a functional active lock-in amplifier, begin with a stable reference source. A Wien-bridge oscillator using an op amp (e.g., LM741 at low frequencies or OPA627 for lower distortion) can generate a clean sine wave. Alternatively, a direct digital synthesis (DDS) module provides precise frequency and phase control. The input signal should first be amplified using a low-noise preamplifier (e.g., with the LT1028) to match the dynamic range of the multiplier. The multiplier stage uses an AD633, whose output is fed into a Sallen-Key low-pass filter with a cutoff frequency of 1 Hz to 100 Hz depending on the desired time constant. The filtered signal then passes through a non-inverting amplifier stage with a gain of 10–100. An additional buffer op amp (e.g., TLV2460) provides a low-impedance output. For quadrature detection, a second channel is built using a 90° phase shifter (an all-pass filter) driving a second multiplier. All components should be placed on a ground-plane printed circuit board with careful layout to minimize parasitic capacitance and noise pickup.

Calibration and Testing

Before use, the lock-in amplifier must be calibrated. Apply a known small AC signal at the reference frequency and adjust the phase of the reference to maximize the output. Record the output voltage for various input amplitudes to verify linearity. To test noise rejection, inject a signal at an adjacent frequency and measure the attenuation. The performance should be validated against a commercial lock-in amplifier if available. For in-field use, periodic calibration is recommended to account for component drift.

Applications

Active lock-in amplifiers built with op amps find applications in many areas. In optics, they are used to measure photocurrents from photodiodes in lock-in configuration to eliminate ambient light noise. In atomic force microscopy, they process the cantilever deflection signal. In biomedical engineering, they extract bioimpedance measurements from electrodes that are prone to interference. They also appear in lock-in thermography, LVDT-based position sensing, and corrosion monitoring. The low cost and customizability make them ideal for educational experiments and prototype development.

Advantages and Limitations

The primary advantages of op amp-based lock-in amplifiers are cost savings, simplicity, and the ability to tailor the design for specific frequency ranges and dynamic reserves. They serve as excellent learning platforms for students and engineers. However, they have limitations. Commercial lock-in amplifiers offer exceptionally low drift, digital signal processing, and very high dynamic reserves (often > 100 dB). Op amp designs may struggle with frequencies above a few hundred kilohertz due to the bandwidth limitations of available multiplier ICs and op amps. Furthermore, analog multipliers introduce nonlinearities that must be compensated. For applications requiring extreme sensitivity or automated measurements, a commercial instrument remains the recommended choice.

Conclusion

Creating active lock-in amplifiers with operational amplifiers is a practical and educational approach to sensitive signal detection. By carefully designing the mixing, filtering, and amplification stages, it is possible to achieve performance suitable for many scientific and industrial applications. The key to success lies in selecting appropriate low-noise op amps, minimizing external interference, and understanding the trade-offs between speed, noise, and dynamic range. For further reading, the application notes from Analog Devices on lock-in techniques and Texas Instruments’ guide to active filters provide excellent reference material. With careful design and testing, an op amp-based lock-in amplifier can become a reliable tool in any signal processing arsenal.