Short answer
A distribution transformer should not be assumed to block every high-frequency disturbance between its voltage levels. Coupling depends on winding and terminal paths, capacitance, grounding, connected impedances and frequency. A fundamental-frequency equivalent may omit the relevant mechanisms. Define the measured frequency band and terminal configuration before interpreting a supraharmonic transfer result or applying it to another installation.
What the published experiment establishes
Slangen and co-authors reported laboratory measurements of transfer through MV/LV transformers in their CIRED 2023 work. The author-hosted abstract describes propagation between voltage levels and amplification of some components in that installation. This establishes that transfer and amplification are possible; it does not provide a universal transfer ratio for distribution transformers.
The abstract describes three transformers of at least 1 MVA. Different transformers, terminations or network impedances can produce different responses. The full proceedings paper was not used here to claim detailed measurement procedures or a general design rule.
Separate modes and physical paths
A differential-mode disturbance between phase conductors is not the same measurement as a common-mode disturbance relative to a protective reference. State which terminal quantities were measured and how the reference was connected. Interwinding, winding-to-ground and external cable paths may participate alongside magnetic coupling.
A delta connection that changes low-order zero-sequence behavior does not by itself prove that high-frequency capacitive coupling is absent. Begin with transformer data for a harmonic network model, then establish whether that model retains the paths needed for the higher-frequency question.
A capacitor-path example
Assumptions. One ideal 1 nF capacitor represents only an illustrative coupling path. Its value is invented for calculation, not taken from a transformer test. Lead inductance, losses and other branches are omitted.
|XC| = 1/(2πfC)
| Frequency | Reactance magnitude |
|---|---|
| 10 kHz | About 15.9 kΩ |
| 100 kHz | About 1.59 kΩ |
The tenfold frequency increase lowers this path’s reactance tenfold. It does not establish the voltage delivered to another winding. That requires the source impedance, receiving impedance and complete network, which can introduce resonances and damping.
Match measurement scope to the claim
The public scope of IEC 61000-4-30:2025, corrected in July 2026, addresses conducted-emission measurements over 2–150 kHz, with separate methods identified for 2–9 kHz and 9–150 kHz. This supplies a measurement framework, not a transformer attenuation requirement. Its full measurement clauses were not audited for this article.
Record frequency, injection and observation sides, operating load, grounding, sensor response and connected cable or filter impedances. Distinguish a measured terminal ratio from a fitted model or an extrapolation. Validate frequency-dependent damping over the band where it is used, rather than extending a low-order resistance assumption indefinitely.
Report the validated band and unresolved sensitivities. Avoid a fixed attenuation, isolation or immunity promise based on a rating label, one laboratory ratio or this ideal-capacitor example.
Sources and scope
- T. Slangen, E. C. W. de Jong, V. Cuk and J. F. G. Cobben, Transfer of Supraharmonics Through a MV/LV Transformer, CIRED 2023, DOI 10.1049/icp.2023.0254. Author-hosted abstract and bibliographic record reviewed.
- IEC 61000-4-30:2025, corrected version 2026-07, and COR1:2026. Official public scope and version records reviewed; full clauses not audited.