Amorphous Toroidal Gap Core: Gapped Ring Cores for High DC Bias Inductors
Sep 30, 2025
Leave a message
What a Gapped Amorphous Ring Core Is
A gapped amorphous toroidal core is a ring wound from amorphous metal ribbon and then cut or otherwise modified to introduce a controlled non-magnetic gap in the magnetic path. The gap is small, often a fraction of a millimetre, but its effect on the B-H curve is large. Without a gap the closed path saturates as soon as a direct current is superimposed on the alternating signal, and inductance collapses. With a gap the slope of the B-H curve becomes nearly constant, so inductance holds up to the design bias current and the stored energy the core can support rises sharply.
The base material is the same Fe-based ribbon used for ungapped amorphous cores: a saturation flux density of 1.56 T, ribbon thickness of 25 µm, hardness around 960 HV, Curie temperature of 410 °C, crystallisation temperature near 500 °C, resistivity of about 130 µΩ·cm, density of 7.18 g/cm³ and saturation magnetostriction of about 27×10-6. Because the ribbon is hard and brittle in bending, the gap must be produced by a controlled process, and the cut faces must be ground and cleaned before assembly.
| Property | Typical value |
|---|---|
| Saturation flux density Bs | 1.56 T |
| Vickers hardness | about 960 HV |
| Curie temperature Tc | 410 °C |
| Crystallisation temperature Tx | about 500 °C |
| Resistivity | about 130 µΩ·cm |
| Density | 7.18 g/cm³ |
| Saturation magnetostriction | about 27×10-6 |
| Adjustable permeability | 50 to 300 |
Gap, Permeability and Inductance
Effective permeability of a gapped ring is set almost entirely by the gap and the geometry, not by the ribbon. The relation is straightforward: total reluctance is the sum of the reluctance of the magnetic path and the reluctance of the gap, so as the gap grows, effective permeability falls and the inductance per turn squared, the AL value, falls with it. Suppliers therefore quote cores by AL value in microhenries per turn squared together with a tolerance, typically plus or minus 15 percent, and by nominal effective permeability.
| Core size od × id × h (mm) | Finished OD × ID × H (mm) | Ae (cm²) | Le (cm) | Permeability | AL at 10 kHz, 0.3 V (µH/N²) |
|---|---|---|---|---|---|
| 22 × 13 × 10 | 24.6 × 10.4 × 13 | 0.39 | 5.5 | 100 | 0.090 |
| 25 × 15 × 15 | 27.6 × 13.3 × 17.2 | 0.65 | 6.28 | 100 | 0.130 |
| 26 × 16 × 10 | 28.2 × 13.9 × 12.7 | 0.43 | 6.59 | 100 | 0.090 |
| 32 × 20 × 15 | 34.5 × 17.8 × 17.8 | 0.77 | 8.16 | 100 | 0.120 |
| 40 × 25 × 15 | 44 × 21.5 × 19 | 0.97 | 10.21 | 100 | 0.120 |
| 40 × 25 × 20 | 45 × 21.5 × 24.7 | 1.29 | 10.21 | 100 | 0.160 |
| 46 × 27 × 15 | 48.5 × 24.4 × 19 | 1.23 | 11.46 | 100 | 0.135 |
| 46 × 27 × 20 | 48.5 × 24.4 × 22.3 | 1.63 | 11.46 | 100 | 0.180 |
| 46 × 27 × 25 | 48.5 × 24.4 × 27.3 | 2.05 | 11.46 | 100 | 0.225 |
The finished dimensions are larger than the wound dimensions because impregnation, the gap and the clamping hardware add material. A designer who dimensioned the bobbin from the wound size alone will not fit the core into the space reserved for it, so the finished outline must be used for mechanical layout.
DC-Bias Behaviour and How to Specify a Gap
Two performance points define a gapped ring: the inductance at zero bias and the DC current at which inductance has fallen by a stated percentage, often 10 percent or 20 percent. Both depend on the gap and on the number of turns. Increasing the gap reduces the small-signal inductance but increases the bias current that can be tolerated before inductance falls, so the gap is effectively a design variable used to trade inductance against current capability.
Because a mechanical gap and an AL value are not perfectly interchangeable across production lots, the specification should state the magnetic requirement, for example the inductance at a given ampere-turn product, and treat mechanical gap as a process parameter. Where the design is sensitive, inductance should be checked at the operating bias rather than at a small-signal level, because a small-signal measurement will report a figure the circuit never sees.
Assembly and Winding Rules
Keep cut faces clean and dry; a trapped particle changes the gap and shifts inductance.
Clamp evenly with a non-magnetic band or a moulded housing rather than a single bolt, which would tilt the gap faces.
Measure inductance after clamping, since clamp pressure changes the effective gap slightly.
Rewind rather than re-gap a core if the target inductance changes; repeated grinding removes material and alters the finished dimensions.
Re-impregnate after any rework to restore mechanical integrity and prevent the ribbon layers from moving.
Typical Applications
Gapped amorphous rings are used as output smoothing chokes in switched-mode power supplies, as inverter chokes in photovoltaic and active filter equipment, as power factor correction inductors and as energy storage inductors in high-current DC links. In all of these the common requirement is a substantial DC component together with a moderate ripple, which is exactly the duty a gapped amorphous ring is designed for. Loss and inductance of finished cores are commonly verified on the toroidal specimen under IEC 60404-6 or by the voltmeter-ammeter-wattmeter method of ASTM A927/A927M, at a stated flux density and frequency.
Frequently Asked Questions
Q: What permeability range can a gapped amorphous core achieve?
A: Effective permeability is adjustable by gap size and is typically supplied in the range of 50 to 300, which corresponds to a wide span of AL values for the same core outline.
Q: How much DC bias can a gapped amorphous ring handle?
A: There is no single figure; it depends on gap size and turns. The core saturates only when the total flux, DC plus ripple, approaches 1.56 T, so a larger gap and more turns allow a higher bias current for the same inductance.
Q: Does gapping increase core loss?
A: Yes, to a degree. Fringing flux near the gap causes local eddy currents and can raise loss near the cut faces, which is why the gap region should be kept away from the hottest part of the winding and why distributed gaps are used in high-ripple designs.
Q: Why is the finished outline larger than the wound core size?
A: Impregnation resin, the machined gap and the clamping hardware all add dimension, so the finished OD, ID and height must be used when designing the bobbin and the surrounding space.
Q: Can inductance be measured at a low signal level for a gapped core?
A: It can, but the result is optimistic. Because permeability varies with flux, the meaningful figure is inductance measured at the actual DC bias and ripple level the inductor will see in service.
Q: Are gapped amorphous cores suitable above 50 kHz?
A: They can operate there, but loss rises with frequency and the winding usually becomes the limiting factor, so above roughly 50 kHz alternative core materials are often evaluated on both loss and cost.