Iron-Silicon-Boron Amorphous Toroidal Core for Transformers and Inductors

Oct 14, 2025

Leave a message

The Iron-Silicon-Boron Ribbon

An iron-silicon-boron amorphous core is wound from ribbon produced by melt spinning, in which molten alloy is ejected onto a rotating copper wheel and solidified at a rate near 106 K/s. The atoms are frozen before they can arrange themselves into a crystal lattice, so the ribbon consists of a single amorphous phase with no grains, no grain boundaries and no magnetocrystalline anisotropy. The practical consequence for a transformer or inductor designer is that magnetisation reversal requires very little energy, and the hysteresis loop is narrow even at flux densities approaching saturation.

Typical properties of the commercial Fe-Si-B ribbon used for these cores are a saturation flux density of about 1.56 T, a Curie temperature of 410 °C, a crystallisation temperature near 500 °C, electrical resistivity between 110 µΩ·cm and 130 µΩ·cm, saturation magnetostriction below 30×10-6, and Vickers hardness around 960 HV. The 25 µm ribbon thickness and the high resistivity together hold eddy-current loss down, which is why the material remains usable well above the frequency where electrical steel laminations become impractical.

Magnetic and Mechanical Characteristics

Property Typical value Design significance
Saturation flux density Bs 1.56 T allows a high flux swing in a small cross section
Ribbon thickness 25 µm low eddy-current component at medium frequency
Stacking factor above 0.8 effective cross section versus measured outline
Curie temperature 410 °C sets the upper service limit well above normal winding temperatures
Resistivity 110 to 130 µΩ·cm eddy-current loss stays low at 10 kHz to 50 kHz
Saturation magnetostriction below 30×10-6 low acoustic noise, but impregnation is still needed
Hardness about 960 HV cores must be handled and clamped without point loading

Two mechanical points are often missed. First, the ribbon is hard and thin, so a core that is dropped will show a chipped or crazed surface and can suffer a measurable loss increase. Second, magnetostriction, although low for an amorphous alloy, is not zero; an unstaked core will buzz in an audible-band design, and proper resin impregnation plus a firm clamp removes most of it.

Loss Behaviour in Transformer and Inductor Service

Total core loss is the sum of hysteresis and eddy-current loss. In amorphous ribbon the hysteresis component dominates at frequencies up to a few kilohertz, and the eddy-current component grows roughly with the square of frequency and the square of flux density. Because the ribbon is only 25 µm thick and its resistivity is roughly five times that of silicon steel, the crossover point where eddy-current loss overtakes hysteresis sits much higher than for a laminated core.

Loss measurement for cores of this geometry is normally done on ring or toroidal specimens. IEC 60404-6 covers the measurement of soft magnetic metallic and powder materials by ring specimens from 20 Hz to 200 kHz, and ASTM A927/A927M covers alternating-current magnetic properties of toroidal cores using the voltmeter-ammeter-wattmeter method. Test data should state flux density, frequency, temperature and whether the sample was impregnated, because winding tension and resin cure both shift measured loss by a few percent.

Winding Practice for Toroidal Cores

A toroid gives a closed magnetic path with very little external field, which is why it is preferred for low-leakage transformers and for inductors in noise-sensitive circuits. To keep that advantage:

Wind uniformly around the full circumference; a bunched winding creates a local flux concentration and extra loss.

Deburr lead exits and use a smooth, radiused inner edge so the ribbon is not cut by the winding.

Keep the winding distributed rather than layered where possible; distributed turns lower interwinding capacitance and reduce the risk of resonant ringing.

Leave the impregnation intact and re-impregnate if the core is machined or gapped after delivery.

Where a DC component is present, an ungapped toroid will saturate quickly because the closed path has no gap to store energy. In that case a gapped toroid or a C core is the correct choice, and the gap should be specified together with an inductance tolerance measured at the operating bias.

Service Conditions and Reliability

Amorphous cores are used in switched-mode power supplies, photovoltaic inverters, electric-vehicle onboard chargers, welding equipment and induction heating supplies. In each case the failure mode to guard against is not demagnetisation but thermal excursion. Amorphous ribbon must not be exposed above about 400 °C in service, and it must never see the 500 °C crystallisation region, because crystallisation permanently destroys the low-loss structure. A winding that runs hot because of copper loss therefore damages the core as well as the insulation, so temperature rise must be checked on the complete assembly rather than on the core alone.

Frequently Asked Questions

Q: What is an iron-silicon-boron amorphous core made of?
A: It is wound from Fe-based amorphous ribbon whose main constituents are iron, silicon and boron, with the balance being minor alloying additions; the alloy is defined by the ribbon supplier rather than by a single fixed composition.

Q: Is an amorphous toroidal core better than a ferrite core at 20 kHz?
A: It offers much higher saturation flux density and better DC-bias tolerance, so it usually wins where size or bias is the constraint; ferrite remains competitive when flux swing is small and cost per piece dominates.

Q: What is the maximum operating temperature of the core?
A: Amorphous Fe-Si-B ribbon has a Curie temperature of about 410 °C and crystallises near 500 °C, but the practical limit is set by the winding insulation class and by the resin used for impregnation, which are far lower.

Q: Which test method applies to a toroidal amorphous core?
A: Ring-specimen measurement from 20 Hz to 200 kHz follows IEC 60404-6, and alternating-current magnetic properties of toroidal cores can be verified by the voltmeter-ammeter-wattmeter method of ASTM A927/A927M at the specified flux density and frequency.

Q: Why does core loss increase after winding?
A: Winding and clamping apply mechanical stress to the ribbon, which raises coercivity and hysteresis loss. Uniform, moderate tension and a firm but not crushing clamp keep the increase small.

Q: Can the core be used in a common-mode choke?
A: Yes, though nanocrystalline ribbon is often preferred there because its much higher permeability gives more impedance at low frequency with fewer turns, while amorphous ribbon is chosen when higher saturation flux density is needed.

Send Inquiry