Development of a Ferromagnetic Resonance Measurement System Using NanoVNA

Ferromagnetic resonance (FMR) is a fundamental technique for probing magnetization dynamics in spintronic and magnetic materials. However, conventional FMR measurements rely on broadband vector network analyzers (VNAs), whose high cost limits accessibility for small laboratories and educational environments. To overcome this barrier, we have developed a compact and low-cost FMR measurement platform - the NanoVNA-FMR system-based on a commercially available NanoVNA. The setup integrates an electromagnet and a coplanar waveguide (CPW) and is fully automated using Python scripts. This enables synchronized magnetic-field sweeping, S-parameter acquisition, and real-time visualization. The system successfully captures clear FMR spectra that exhibit systematic shifts in resonance frequency with increasing magnetic field. The results are in excellent agreement with those obtained using a conventional VNA-based FMR system, confirming the quantitative reliability of the NanoVNA approach. Additionally, a 3D-printed sample holder further reduces overall system cost. These results demonstrate that the NanoVNA-FMR system provides a practical, accurate, and accessible alternative for quantitative magnetic characterization and educational applications.

💡 Research Summary

The paper presents the design, implementation, and validation of a low‑cost ferromagnetic resonance (FMR) measurement platform built around a commercially available NanoVNA (NanoVNA‑F V2). Conventional broadband FMR setups rely on high‑end vector network analyzers (VNAs) that cost tens of thousands of dollars, limiting their accessibility to well‑funded laboratories and making them impractical for teaching environments. The authors address this limitation by exploiting the NanoVNA’s capability to generate and measure S‑parameters up to 3 GHz, its USB interface, and its modest price (a few hundred dollars).

The experimental configuration integrates the NanoVNA with a high‑performance coaxial cable (Thorlabs SMM‑36), a 50 Ω coplanar waveguide (CPW) fabricated by Hayashi Repic Co., a C‑frame DC electromagnet (TESLA TMSP232‑1204015) capable of producing up to ~45 mT in the plane of the sample, and a programmable DC power supply (Owon SPE6103) that drives the electromagnet with a calibrated conversion of 1 V → 2.5 mT. A custom 3D‑printed sample holder made of PLA secures the YIG thin‑film specimen (≈2 µm thick, 3 × 5 mm²) on the CPW central conductor while minimizing vibration and misalignment.

Automation is achieved through a Python script that simultaneously controls the NanoVNA and the DC power supply. The script performs the following sequence: (i) initialize the NanoVNA to sweep 50 kHz–3 GHz, (ii) record a reference S21 spectrum at 0 mT, (iii) step the magnetic field from 0 to 45 mT in 2.5 mT increments (0.5 V steps), (iv) acquire S21 at each field, (v) compute differential transmission ΔS21(dB) = S21(H) − S21(0 mT), (vi) fit each ΔS21 curve with a Lorentzian function to extract the resonance frequency and linewidth, and (vii) store all data as CSV files while providing real‑time plots of the spectra, peak positions, and a two‑dimensional frequency‑field colormap.

Measurements on a YIG/GGG sample reveal clear resonance peaks that shift to higher frequency with increasing magnetic field, as expected for in‑plane magnetization. Lorentzian fits yield resonance frequencies that follow the Kittel equation f = (γ/2π)·√