A physics magnet in an electrical context is a material that generates a persistent magnetic field or concentrates an applied magnetic flux, governed by its non-linear B-H (magnetic flux density versus magnetic field strength) curve. When you wrap wire around a magnetic core and push alternating or pulsed current through it, the material does not just scale linearly forever. In a real circuit, the choice of magnetic core changes the inductance, energy storage capacity, and high-frequency heat dissipation (core losses), ultimately dictating whether a power supply runs efficiently or fails catastrophically.

The One-Sentence Definition and the B-H Curve

To understand applied physics magnets on the workbench, you have to look at the B-H curve. This graph plots H (the magnetic field strength, measured in Amperes per meter, which is directly proportional to your current) on the X-axis, and B (the magnetic flux density, measured in Tesla, which is the actual magnetic flux inside the core) on the Y-axis.

At low currents, the relationship is roughly linear. The slope of this line is the material's permeability ($\mu$). But as H increases, the curve bends and eventually flattens out. This flattening is core saturation. Once a physics magnet saturates, it cannot hold any more magnetic flux. The permeability drops to near that of free space (air), and the inductor effectively stops acting like an inductor.

Bench Rule of Thumb: An inductor is only an inductor as long as its core is unsaturated. Once saturation hits, it becomes a very low-value resistor, limited only by the DC resistance (DCR) of the copper wire.

What People Commonly Confuse About Physics Magnets

The most common mistake hobbyists and junior engineers make is confusing H (effort) with B (result). Think of H as the number of cars entering a highway (current), and B as the actual traffic density moving through (flux). Once the highway is jammed (saturation), adding more cars (current) just creates heat and gridlock, not more forward movement.

Another major confusion is assuming a core's permeability is a fixed, constant number. Datasheets list 'initial permeability' ($\mu_i$), but this is only valid at very low signal levels (usually measured at < 10 A/m). Under real power conditions, permeability shifts with temperature, DC bias current, and frequency. If you design a circuit using only the initial permeability number without checking the DC bias curves, your design will likely fail under load.

Worked Numeric Example: Sizing a Ferrite Core Inductor

Let's run the numbers for a switch-mode power supply inductor. We need 100 µH of inductance, and the peak current ($I_{pk}$) will be 5A. We select a standard TDK N87 ferrite toroid.

  • Core Parameters (from datasheet): $A_L$ value = 250 nH/turns², Effective magnetic path length ($l_e$) = 50 mm (0.05 m), Effective cross-sectional area ($A_e$) = 100 mm² ($1 \times 10^{-4}$ m²).
  • Turns Calculation: $L = N^2 \times A_L$. Therefore, $100,000 nH = N^2 \times 250 nH$. $N^2 = 400$, so N = 20 turns.

Now, we must verify that 5A won't saturate the core. We calculate the peak flux density ($B_{pk}$) using the formula:

$B = (L \times I) / (N \times A_e)$

$B_{pk} = (100 \times 10^{-6} H \times 5 A) / (20 \text{ turns} \times 1 \times 10^{-4} m^2) = \mathbf{0.25 \text{ Tesla}}$

N87 ferrite saturates at approximately 0.39T at 100°C. Because our calculated 0.25T is well below the 0.39T saturation threshold (leaving a ~35% safety margin), this core will not saturate at 5A peak. The design is safe. For a deeper look at how hysteresis loops dictate these limits, refer to the All About Circuits guide on magnetic hysteresis.

Where You Meet This in Practice

You interact with the physics of magnetic cores every time you work with alternating or switching currents. Here is where the B-H curve dictates hardware behavior:

  1. Mains Transformers (50/60Hz): These use silicon steel laminations. Steel has a high saturation point (~1.5T to 2.0T) but high hysteresis loss at high frequencies. If you apply 50kHz to a silicon steel transformer, the hysteresis loop area (energy lost as heat per cycle) will cause the core to melt.
  2. DC-DC Converter Inductors: These use ferrite or iron powder cores. Ferrites have low saturation points (~0.3T to 0.4T) but incredibly low high-frequency losses. They must be physically 'gapped' (a slice of air cut into the magnetic path) to artificially lower the permeability and increase the current required to reach saturation.
  3. Relays and Contactors: The difference between the 'pull-in' voltage and the much lower 'drop-out' voltage of a relay coil is a direct result of magnetic reluctance and the hysteresis of the soft iron core. The physics magnet wants to stay magnetized even after the coil is de-energized.

Real-World Scenario Walkthrough: The Blown Buck Converter

Theory is clean; the workbench is not. Here is a classic failure mode involving physics magnets that I see constantly in DIY power electronics.

The Setup: A maker is building a 12V to 5V buck converter switching at 100 kHz. The design requires a 47 µH inductor. Looking through their parts bin, they find an un-gapped FT-50 ferrite toroid (Material 43) and wind 15 turns of 20 AWG wire.

The Numbers: Material 43 has a high $A_L$ of roughly 420 nH/turns². The LCR meter, which tests at 1mA, reads $15^2 \times 420 = 94.5 \mu H$. The maker thinks, "Great, more inductance than I need, the ripple current will be tiny." The target peak current in the circuit is 3A.

The Outcome: The circuit powers up fine at no load. But the moment a 3A load is connected, the MOSFET switch violently explodes, taking the driver IC with it.

What Went Wrong: The un-gapped ferrite core hit hard saturation at roughly 0.5A. At 3A, the core was completely saturated, and the inductance collapsed from 94.5 µH down to the 'air core' value of about 0.05 µH. The rate of current rise is dictated by $di/dt = V/L$. With L dropping by a factor of nearly 2000, the current spiked to tens of amps in a single microsecond. The MOSFET exceeded its 10A absolute maximum rating before the control IC could even finish its current switching cycle.

The Fix: Never use un-gapped high-permeability ferrite for DC-biased power inductors. Use a distributed-gap iron powder core (like a Micrometals/TDK T50-26) or a specifically gapped ferrite core. These materials exhibit a 'soft' saturation curve, meaning inductance rolls off gradually rather than dropping off a cliff. Check the Würth Elektronik magnetics catalog for pre-gapped power inductors rated for specific DC bias currents.

FAQ: Physics Magnets in Circuit Design

Why do transformer cores get hot even with no electrical load connected?

This is caused by two physics magnet phenomena: hysteresis loss and eddy currents. Even with no secondary load, the primary coil draws a 'magnetizing current' that constantly reverses the magnetic domains in the core. The friction of these domains flipping back and forth generates heat (hysteresis). Simultaneously, the changing magnetic field induces tiny circulating currents inside the conductive core material itself (eddy currents), which generate $I^2R$ heating. This is why high-frequency cores are made of non-conductive ferrite ceramics or insulated steel laminations.

What is the difference between 'hard' and 'soft' physics magnets?

In magnetics, 'hard' refers to materials with high coercivity—they resist being demagnetized and are used for permanent magnets (like Neodymium in BLDC motors). 'Soft' refers to materials with low coercivity—they magnetize and demagnetize easily with very little energy loss. Transformer cores and inductors strictly use soft magnetic materials. If you accidentally use a hard magnetic material for an AC transformer core, the massive hysteresis loop will cause it to overheat and fail in seconds.

How does temperature affect core saturation?

For most standard ferrites (like N87 or Material 43), the saturation flux density decreases as temperature rises. A core that safely handles 0.35T at 25°C might saturate at 0.30T when it heats up to 100°C inside an enclosed project box. Always derate your maximum flux density target by at least 20% to account for operating temperature rise.