A magnetist (commonly known as a magnetics engineer) applies the principles of electromagnetism to design inductors, transformers, and chokes that store and transfer energy in electrical circuits. In a real power installation or PCB, their design choices dictate the physical size, switching frequency limits, thermal efficiency, and electromagnetic interference (EMI) profile of the entire system. Beginners commonly confuse magnetic flux density (B) with magnetic field strength (H), or mistakenly assume inductance is a fixed, immutable property rather than a variable that collapses when a core saturates.
Core Material Selection: The Magnetist's Palette
The foundation of any magnetic component is the core material. A magnetist doesn't just grab a random chunk of iron; they select materials based on permeability, saturation limits, and core loss characteristics at specific switching frequencies. Below is a reference table of the most common core materials used in modern power electronics, from low-frequency line transformers to high-frequency switch-mode power supplies (SMPS).
| Material Type | Initial Permeability ($\mu_i$) | Saturation Flux ($B_{sat}$) | Frequency Range | Primary Application |
|---|---|---|---|---|
| MnZn Ferrite (e.g., TDK N87) | 2,000 - 3,000 | ~0.39 T (at 100°C) | 10 kHz - 500 kHz | SMPS transformers, high-power inductors |
| NiZn Ferrite | 10 - 2,000 | ~0.30 T | 1 MHz - 100 MHz | EMI suppression beads, RF chokes |
| Powdered Iron (e.g., Micrometals -26) | 75 | ~1.2 T (soft roll-off) | DC - 100 kHz | PFC chokes, continuous-mode buck inductors |
| Nanocrystalline (e.g., Metglas) | 10,000 - 100,000 | ~1.2 T | 50 Hz - 20 kHz | High-efficiency 50/60Hz line transformers, CTs |
Notice the trade-off: materials with high saturation limits (like powdered iron) generally have low permeability, meaning you need more turns of wire to achieve the same inductance. Conversely, high-permeability MnZn ferrites saturate easily and require careful gap calculations or strict current limiting. For a deeper look at how these materials behave under alternating fields, review the B-H curve and hysteresis loops explained in standard circuit theory texts.
The Math in Practice: Sizing an Inductor Without Saturating
Let's move from theory to the workbench. Suppose you are designing a buck converter and need a 1 mH (1000 $\mu$H) inductor that can handle a peak current of 0.5 A without saturating. You choose a standard TDK ETD39/20/13 core with N87 ferrite material.
First, we look up the core's $A_L$ value (inductance per turn squared) from the manufacturer's design tools or datasheet. For an ungapped ETD39 N87 core, $A_L$ is approximately 2600 nH/N² (or 2.6 $\mu$H/N²). The effective cross-sectional area ($A_e$) is 125 mm² ($1.25 \times 10^{-4}$ m²).
Step 1: Calculate the required turns (N).
The formula is $L = A_L \times N^2$. Rearranging for N:
$N = \sqrt{\frac{L}{A_L}} = \sqrt{\frac{1000 \mu H}{2.6 \mu H}} = \sqrt{384.6} \approx 19.6$
We round up to 20 turns. With 20 turns, our actual inductance is $20^2 \times 2.6 = 1040 \mu H$ (1.04 mH).
Step 2: Verify the peak flux density ($B_{max}$).
This is where many hobbyists fail, resulting in a melted MOSFET when the core saturates. The formula is:
$B_{max} = \frac{L \times I_{peak}}{N \times A_e}$
Plugging in our real values:
$B_{max} = \frac{1.04 \times 10^{-3} H \times 0.5 A}{20 \times 1.25 \times 10^{-4} m^2} = \frac{0.00052}{0.0025} = 0.208 Tesla$
Because N87 ferrite saturates at roughly 0.39 T at 100°C, our calculated 0.208 T leaves a safe ~46% margin. If we had pushed the peak current to 2.0 A, $B_{max}$ would hit 0.83 T, the core would saturate, inductance would drop to near zero, and the switching transistor would instantly fail from overcurrent. This calculation is the daily bread of a working magnetist.
Where You Meet This in Practice
You interact with the results of magnetics engineering every time you plug in a modern electronic device. Here is where these components live in the wild:
- Switch-Mode Power Supplies (SMPS): In a flyback converter (like your laptop charger), a coupled inductor stores energy in the core's magnetic field during the MOSFET's 'on' time and transfers it to the secondary winding during the 'off' time. The magnetist must calculate the exact air gap to store energy without saturating.
- Resonant Converters (LLC): Used in high-efficiency server power supplies, these rely on the transformer's leakage inductance and magnetizing inductance to achieve Zero Voltage Switching (ZVS). The physical spacing between primary and secondary windings is tightly controlled to tune this leakage.
- EMI Filtering: Common-mode chokes use high-permeability toroidal cores to present high impedance to high-frequency noise while allowing 50/60Hz line current to pass unimpeded.
- High-Frequency Winding Losses: At frequencies above 50 kHz, the skin effect forces current to the outer edge of the wire, and the proximity effect causes currents in adjacent winding layers to crowd together. To solve this, magnetists specify Litz wire (many individually insulated thin strands woven together) to maximize the effective surface area for high-frequency RMS currents.
Common Confusions in Magnetic Theory
When troubleshooting or designing magnetic circuits, keep these distinctions clear to avoid costly design errors:
Inductance vs. Impedance
Inductance (measured in Henries) is a physical property of the component based on its geometry and core material. Impedance (measured in Ohms) is how much that inductor resists alternating current at a specific frequency ($X_L = 2\pi fL$). A 10 $\mu$H inductor has the same inductance at 1 kHz and 1 MHz, but its impedance is 628 times higher at 1 MHz.
Flux Density (B) vs. Field Strength (H)
Field strength (H, measured in Amperes per meter) is the effort you put in, driven purely by the current and the number of turns ($H = \frac{N \times I}{l_e}$). Flux density (B, measured in Tesla) is the result you get out, which depends on the core material's permeability ($B = \mu \times H$). If you swap an air core for a ferrite core, H stays exactly the same for a given current, but B increases by a factor of 2,000 or more.
Core Loss vs. Copper Loss
Heat in an inductor comes from two places. Copper loss ($I^2R$) is resistive heating in the wire windings. Core loss is heating inside the magnetic material itself, caused by hysteresis (the energy required to constantly flip the magnetic domains back and forth) and eddy currents. A magnetist balances these by selecting wire gauge for copper loss and core material/geometry for core loss, often targeting a 50/50 split for optimal thermal performance.
