Measuring harmonics in practice — power quality analyzer, THD-I/THD-V and K-factor transformer derating
Measuring harmonics in practice — power quality analyzer, THD-I/THD-V and K-factor transformer derating
The guide on harmonics & THD limits covers the THD requirement from the Netcode Elektriciteit for large consumers. This article goes one step further: how is harmonic distortion actually measured in practice, and what do you do with that measurement when an existing transformer is involved?
THD-I versus THD-V: why current is usually the starting point
THD-V (voltage THD) describes how distorted the voltage waveform is at a measurement point — this is the result of the combined loading of the entire network upstream of that point, and does not directly indicate which individual load causes the distortion. THD-I (current THD) describes the distortion of the current drawn by a specific load — a variable frequency drive, LED driver or switch-mode power supply generates high THD-I regardless of how clean the supply voltage is. When tracing the source of harmonic distortion in an installation, THD-I is therefore usually the more useful starting point: it points directly to the load distorting the current waveform, whereas THD-V mainly says something about the resulting effect on the whole network.
Measurement method under IEC 61000-4-7
IEC 61000-4-7 sets out the measurement method for harmonics and interharmonics:
- a 200 ms measurement window (10 cycles at 50 Hz), within which each harmonic component is determined via Fourier analysis;
- harmonics up to a high order (typically up to the 40th or 50th) are reported individually, alongside the summarised THD value;
- a Class A power quality analyzer (per IEC 61000-4-30) is required for contractual or dispute-resolution measurements, where measurement accuracy and repeatability between different instruments must be guaranteed; a simpler Class S instrument suffices for an exploratory survey of an installation.
In practice, a clamp-type power quality analyzer (for example a Fluke 435 series or comparable) is clamped around each phase conductor — and, in a three-phase four-wire system, around the neutral conductor as well — after which the instrument reports the current and voltage harmonics per phase.
K-factor: what it says about an existing transformer
An ordinary transformer is designed for a purely sinusoidal load current (K = 1). Loads that draw harmonic current (variable frequency drives, LED drivers, switch-mode power supplies) cause additional heat losses in the transformer windings — particularly eddy-current losses, which increase disproportionately with the order of the harmonic. The K-factor weights each harmonic current component by the square of its order, expressing how much extra eddy-current loss a given load current causes in a transformer compared with a purely sinusoidal current of the same RMS value. K-rated transformers (for example K-4, K-13, K-20) are specifically designed to withstand a higher K-factor load without additional derating, per IEEE C57.110.
Practical application: comparing the measured K-factor with the nameplate
For an existing transformer feeding a growing number of variable frequency drives, LED lighting or other non-linear loads, the practical question is not only whether the rated kVA loading is exceeded, but also whether the actual K-factor of the load current stays within the transformer's K-rating. A transformer that stays comfortably within its kVA margin can still overheat structurally if the measured K-factor of the load exceeds the transformer's nameplate value — a situation that an ordinary current or power measurement does not reveal, but a harmonics measurement with K-factor calculation does.
Practical relevance
For a complaint about a transformer running warmer than expected without the rated load being exceeded, or when expanding an installation with many variable frequency drives or LED lighting, measuring THD-I and the K-factor of the load current — and comparing that with the K-rating on the transformer's nameplate — is a more direct diagnosis than simply reading the current in amps.
Common mistakes
- Measuring only voltage THD (THD-V) to trace the source of distortion — THD-V says something about the resulting effect on the network, not which load causes the distortion.
- Using a Class S instrument for a measurement that must serve as evidence in a contractual or dispute context — that requires a Class A analyzer per IEC 61000-4-30.
- Not comparing the transformer's nameplate K-factor with the measured K-factor of the actual load — a transformer without a K-rating (K=1) fed with a high-K-factor load can overheat despite an apparently comfortable kVA margin.
- Not measuring the neutral conductor in a three-phase four-wire system with significant third-harmonic loading — see the guide on the neutral conductor under third-harmonic loading for why that specifically can be a problem.
Related
Further reading
- PracticalHeat pump — electrical connection in practice
- §643 / IEC 60364-6Checking PEN continuity in practice — measurement method and the pitfall of a misleadingly low reading
- PracticalPerilex, CEEform and phase rotation — recognising the wiring
- §531 / praktijkRCBO versus separate RCD + MCB — the practical trade-off
- PracticalReading a distribution panel schema — Eaton & Hager example
- PracticalUpgrading a connection — the application procedure with the grid operator