Dynamic magnetic loss measurement quantifies the energy dissipated in soft magnetic materials when they are magnetized under alternating or pulsed conditions, expressed as specific total loss in watts per kilogram. Accurate determination of dynamic losses is essential for evaluating electrical steel, amorphous alloys, soft ferrites, and powder cores used in transformers, motors, and high-frequency power electronics. This article explains the physical origins of dynamic magnetic losses, the main measurement standards and methods, typical sample requirements, and answers to frequently asked questions.
Technical Overview
Dynamic magnetic loss, often referred to as AC core loss or iron loss, is the energy converted into heat within a magnetic material when it is repeatedly magnetized by an alternating, rotating, or pulsed magnetic field. Unlike static (DC) hysteresis loss, which depends only on the area of the quasi-static hysteresis loop, dynamic loss rises steeply with magnetizing frequency and flux density, making it the dominant efficiency-limiting factor in most electrical machines and power conversion devices.
According to the classical statistical loss theory, the total specific loss Ps (W/kg) measured under sinusoidal flux conditions can be decomposed into three components:
- Hysteresis loss (Ph): energy dissipated as domain walls pin and unpin during each magnetization cycle; proportional to frequency (f).
- Eddy current loss (Pe): Joule heating induced by circulating currents that oppose flux changes; proportional to the square of frequency (f²) and to the square of thickness for laminated materials.
- Excess (anomalous) loss (Pexc): additional dynamic loss associated with non-uniform domain wall motion; approximately proportional to f1.5.
The total loss is commonly expressed as Ps = kh·f·B²max + ke·f²·B²max + kexc·f1.5·B1.5max, where Bmax is the peak magnetic flux density. Because eddy current and excess losses scale so strongly with frequency, dynamic magnetic loss measurement must be performed at the actual operating frequency and waveform of the intended application, using standardized excitation and waveform control.
Dynamic loss characterization is typically performed by accredited third-party laboratories in accordance with the IEC 60404 series, ASTM A804/A343 family, and corresponding GB national standards, providing traceable loss values for material qualification, design input, and regulatory compliance such as energy efficiency class verification for electrical steel.
Testing Samples
Grain-Oriented Electrical Steel (GO)
High-permeability grain-oriented silicon steel strips cut along or across the rolling direction, tested in Epstein frames or single sheet testers. Typical thicknesses range from 0.18 mm to 0.35 mm, with reference loss values commonly reported at 1.7 T and 50/60 Hz.
Non-Oriented Electrical Steel (NO)
Isotropic silicon-iron sheets used in rotating machines, tested in both rolling and transverse directions with L/R/T specimen configurations. Thicknesses from 0.20 mm to 0.65 mm are evaluated at 1.5 T or 1.7 T and 50/60 Hz, and increasingly at 400 Hz to 1 kHz for high-speed motor applications.
Amorphous and Nanocrystalline Ribbons
Rapidly quenched foils (typically 20–30 µm thick) wound into toroidal cores or cut into strip specimens. These low-loss materials are tested at frequencies from 50 Hz up to tens of kilohertz, following procedures adapted from IEC 60404-6 and ASTM A932.
Soft Ferrite Cores
Mn-Zn and Ni-Zn ferrite toroids, E-cores, and pot cores used in switched-mode power supplies, tested between 10 kHz and 1 MHz or higher under controlled sinusoidal flux excitation in line with the IEC 62044 series.
Powder and Composite Cores
Iron powder, sendust (Fe-Si-Al), MPP (Mo-permalloy), high-flux, and soft magnetic composite (SMC) ring cores, evaluated for distributed air-gap inductor and motor applications across a wide frequency spectrum.
Nickel-Iron and Cobalt-Iron Alloys
Permalloy, supermalloy, and Fe-Co-V strips wound as toroids or Epstein specimens, characterized at power and audio frequencies for precision transformer and magnetic shielding applications.
Testing Items
A comprehensive dynamic magnetic loss determination typically includes the following measurands:
- Specific total loss (Ps, W/kg): the core quantity, measured at specified peak flux density (e.g., 1.0, 1.5, 1.7, 2.0 T) and frequency (e.g., 50 Hz, 60 Hz, 400 Hz, 1 kHz, 10 kHz).
- Specific apparent power (Ss, VA/kg): the total volt-ampere burden per unit mass required to sustain the magnetization.
- Specific reactive power (Qs, var/kg): the quadrature component related to stored magnetic energy.
- Dynamic hysteresis loop (B–H loop): the AC loop area, from which loss per cycle is derived, together with coercivity and remanence under dynamic conditions.
- AC amplitude permeability and peak permeability: the ratio of peak flux density to peak field strength under sinusoidal flux.
- Loss separation analysis: decomposition into hysteresis, classical eddy current, and excess loss components via multi-frequency measurements.
- Loss temperature dependence: repeat measurements at controlled ambient temperatures (e.g., 23 °C, 100 °C) to capture thermal sensitivity of ferrites and coated powders.
- Waveform dependence studies: comparison of losses under sinusoidal, trapezoidal, and PWM-type flux excitation relevant to inverter-fed machines.
- Magnetic property curves: loss versus frequency, loss versus flux density, and permeability versus frequency characteristics for design engineers.
Testing Methods
Epstein Frame Method
The classical reference technique defined in IEC 60404-2 and ASTM A804 uses a square frame with primary (magnetizing) and secondary (measuring) windings into which strips of the material are inserted. Air-flux compensation is applied, and loss is determined by the wattmeter or digital sampling method at controlled sinusoidal flux. The Epstein method remains the legal reference for grading electrical steel grades and loss certification.
Single Sheet Tester (SST) Method
Described in IEC 60404-3, GB/T 13789, and ASTM A804-related procedures, the SST measures a single sheet placed in a magnetic yoke, requiring far less material than the Epstein frame. With proper mutual inductance compensation and yoke correction factors, SST results are traceable to Epstein values, enabling rapid routine quality control in steel production lines.
Ring (Toroidal) Specimen Method
Specified in IEC 60404-6, IEC 62044, and GB/T 21219 for isotropic and wound materials, the ring method uniformly winds primary and secondary coils directly onto a toroidal core. The absence of an air gap makes it the preferred geometry for amorphous, nanocrystalline, ferrite, and powder cores, valid from power frequency up to the megahertz range with appropriate winding and instrumentation design.
Wattmeter and Digital Sampling Techniques
Modern systems replace analog wattmeters with high-precision digitizers that simultaneously sample secondary voltage and primary current. Digital integration of the flux waveform, air-flux correction, and FFT-based waveform control ensure that the form factor of the secondary induced voltage is maintained at 1.111 (pure sinusoidal flux), as required by the standards. Loss is computed as the average product of magnetizing current and induced voltage per cycle, divided by specimen mass.
Calorimetric Method
For high-frequency or non-sinusoidal excitation where electrical methods suffer from phase errors, calorimetry measures the temperature rise of the core in an adiabatic enclosure. The dissipated power is derived from the heat balance, offering an inherently direct loss determination independent of waveform distortion, though at the cost of longer test cycles.
High-Frequency and Pulsed Excitation Methods
For ferrites and thin foils at 10 kHz–1 MHz, resonant drive circuits and capacitor discharge methods generate the required excitation, with losses obtained either electrically or calorimetrically. Waveform-capture instruments record B(t) and H(t) under square-wave or PWM flux, allowing engineers to correlate laboratory results with real converter operating conditions.
Quality Assurance and Traceability
All measurement chains should be calibrated against national metrology institute standards, with uncertainty budgets covering voltage and current transducers, digitizer resolution, mass determination, specimen dimensional tolerance, and waveform fidelity. Interlaboratory comparisons among accredited third-party laboratories validate the consistency of reported loss values.
Applications
Dynamic magnetic loss data directly influence the design, efficiency certification, and lifetime economics of a broad range of electrical equipment:
- Power and distribution transformers: grading grain-oriented steel and amorphous cores for no-load loss guarantees and energy efficiency regulations.
- Electric motors and generators: selecting non-oriented steel grades for traction motors, industrial drives, and high-speed machines, where iron loss governs thermal margins.
- Switched-mode power supplies and inverters: choosing ferrite and powder core materials for transformers and output inductors operating at 20 kHz–1 MHz.
- Electric vehicle power electronics: validating magnetic components under PWM excitation and elevated temperatures typical of automotive duty cycles.
- Renewable energy systems: optimizing inductor and transformer cores in photovoltaic converters and wind turbine generators for reduced losses and higher power density.
- Material development and quality control: comparing loss performance of next-generation nanocrystalline, SMC, and thin-gauge steels during research, production, and incoming inspection.
- Energy efficiency compliance: providing the certified loss values needed for efficiency class labeling of electrical steel and finished apparatus under international efficiency programs.
FAQ
What is the difference between dynamic and static magnetic loss?
Static (DC) hysteresis loss is measured under very slow, quasi-static magnetization and reflects only domain wall pinning. Dynamic magnetic loss is measured under AC excitation and additionally includes eddy current loss and excess loss, both of which grow with frequency. Consequently, dynamic loss is always greater than static loss for the same peak flux density.
Why must the flux waveform be controlled during measurement?
Loss depends strongly on the shape of the flux waveform, not just its amplitude and frequency. Standards therefore require sinusoidal flux, verified by a secondary voltage form factor of 1.111. Modern digital feedback systems iteratively correct the primary current waveform until this condition is met, ensuring comparability between laboratories.
Which standards govern dynamic magnetic loss measurement?
Key references include IEC 60404-2 (Epstein frame), IEC 60404-3 (single sheet tester), IEC 60404-6 (ring specimens for soft magnetic materials), the IEC 62044 series (ferrite core losses), ASTM A804 and ASTM A343 for AC magnetic properties, and corresponding GB national standards. The applicable document depends on material type, specimen geometry, and frequency range.
What specimen preparation is required?
Epstein testing requires strips of standard length and width, half cut parallel and half perpendicular to the rolling direction for non-oriented grades. Ring specimens must have a precise ratio of outer to inner diameter to keep demagnetizing effects negligible, uniform winding distribution, and accurate mass and density determination. Stress-relief annealing may be applied to define the reference metallurgical state.
What frequency and flux density ranges can be tested?
Conventional power-frequency measurements cover 50/60 Hz at 1.5–1.7 T for electrical steel. Medium-frequency testing extends to 400 Hz–10 kHz for high-speed machine steels, while ferrite and thin-ribbon characterization reaches 100 kHz–1 MHz and beyond, with flux densities typically limited to 100–300 mT at high frequency to avoid core heating.
How is measurement uncertainty evaluated?
Uncertainty contributions include current and voltage transducer accuracy, phase error between channels, digitizer resolution, air-flux compensation residuals, specimen mass and cross-sectional area tolerances, and waveform fidelity deviation. A comprehensive budget typically yields expanded uncertainties in the range of 1–3 percent for power-frequency loss measurements at accredited laboratories.
Can losses under non-sinusoidal or PWM excitation be measured?
Yes. Programmable power amplifiers with digital waveform synthesis allow application of arbitrary flux waveforms, including trapezoidal and PWM patterns encountered in inverter-fed machines. The measured loss increase relative to sinusoidal excitation provides essential correction factors for machine designers, and calorimetric methods may be used when electrical phase accuracy becomes limiting.