Are You Leaving Phase Information on the Table? Choosing the Right Lock-in Strategy for Your Microwave Measurement

You've spent weeks optimising your microwave setup, only to find your results are noisy and your phase is drifting, and recalibration is bringing further delays. The culprit might not be your sample; it could be your choice of detection mode. 

When characterizing samples and devices at microwave frequencies, choosing the right lock-in detection technique is crucial. This blog explores two common approaches using lock-in amplifiers: down-mixing combined with low-frequency lock-in detection, and direct lock-in measurement at microwave frequencies. We’ll outline when each technique is most appropriate, and how to navigate the trade-offs between simplicity, phase sensitivity, bandwidth, and calibration effort.

Excitation vs. Measurement Frequency: Not Always the Same

Before we begin, a fundamental distinction to keep in mind is between the excitation frequency – the frequency used to stimulate your device – and the measurement (or readout) frequency, where the response is detected.

In photonics, for example, a sample might be excited by light at hundreds of terahertz. However, if the light’s intensity is modulated at 1 kHz, the detector picks up the response at that low modulation frequency. The same principle holds for microwave applications: a device might be driven at 5 GHz but modulated at 1 kHz, and the response of interest is extracted at 1 kHz.

When to Use a Low-Frequency Lock-In Amplifier With Down-Mixing

When the signal of interest evolves slowly (e.g., kHz timescales) and phase information at the microwave carrier is not critical, a low-frequency lock-in amplifier such as the MFLI offers a robust and cost-effective solution. Here’s a typical workflow:

• Apply a low-frequency amplitude modulation (e.g., 1 kHz) to your RF source.
• Use that modulation frequency as a reference input to the lock-in amplifier.
• Measure the amplitude of the response at the modulation frequency.

This method is ideal when you only need amplitude information, such as tracking the resonance of a cavity or temperature-induced shifts, without needing to preserve phase.

In this configuration, the RF generator handles the modulation, and the lock-in amplifier reads out the low-frequency response directly. It’s straightforward, reliable, and effective for many classical applications. IQ mixing is unnecessary, as phase extraction was not required. For applications where the phase is required, read on to learn about IQ mixers can help.

IQ Mixing and Their Drawbacks

An IQ mixer downconverts an RF signal by splitting it into in-phase (I) and quadrature (Q) paths and mixing them with 0°/90° local-oscillator tones. After low-pass filtering, the I and Q outputs form a complex baseband (or low-IF) signal that preserves amplitude and phase for coherent demodulation and flexible DSP. This complex representation distinguishes positive and negative frequencies, providing inherent image rejection compared with single-ended mixers.

IQ mixers are a common tool for downconverting high-frequency signals to baseband, enabling phase and amplitude extraction. However, they come with trade-offs:

• Calibration required: IQ mixers are sensitive to drift, requiring regular recalibration to maintain phase accuracy.
• Artifacts and spurs: Imperfect isolation can lead to signal artifacts that compromise measurement fidelity.
• Bandwidth constraints: High-quality mixers that support hundreds of MHz of sideband spacing are expensive and bulky.
• Independent measurement of sidebands are not possible.

For applications requiring high fidelity, fast feedback, or wideband demodulation, direct microwave lock-in instruments avoid these limitations altogether.

When Direct Lock-in at Microwave Frequencies Becomes Essential

There are scenarios where you need to go beyond simple amplitude tracking:

• Phase-sensitive applications, such as phase-locked loops (PLLs) or Pound-Drever-Hall (PDH) locking.
• Time-domain measurements, such as detecting fast pulses or transients.
• Sideband-selective measurements, especially when sidebands are spaced hundreds of MHz apart and need to be measured independently.

In these cases, preserving the full amplitude and phase information at the microwave frequency is essential, and this is where direct GHz lock-in amplifiers like the SHFLI or GHFLI come into play.

These instruments bypass the limitations of IQ mixing by using double-superheterodyne (DSH) conversion internally, removing the burden from the user. This technique enables:

• Phase-preserving demodulation without the need for manual IQ calibration.
• Wide instantaneous bandwidth (up to 1 GHz).
• Robust operation in multi-frequency and high-speed scenarios.

Summary: When to Use What

Use a low-frequency lock-in with down-mixing (e.g., MFLI) when:

• The signal evolves slowly (low-kHz bandwidth).
• Phase is not essential.
• You want a low-cost solution and don't mind additional set-up complexity.

Use direct RF lock-in (e.g., SHFLI or GHFLI) when:

• Phase accuracy at GHz matters (e.g., for PLL or PDH).
• You need high bandwidth for fast pulses or scanning probe microscopy.
• You're measuring widely spaced sidebands or operating multiple frequencies simultaneously.
• You want calibration-free, high-fidelity operation.

Not sure which technique to use? Our application scientists are happy to engage with you to discover which method is best for your experiment. Get in touch or have a look at the product pages for the MFLI, GHFLI or SHFLI.