Inductors and Their Applications — The Coil, The Myth, The Energy Bank

"If Transformers are the social butterflies of magnetic components (sharing energy between windings), inductors are the lone monks — hoarding energy in their field and doling it out when the circuit begs."

You're already warmed up from the Transformers chapter, and you've been flirting with Control Strategies (yes, MPC stalks the future of inductor currents). Now we dive into the beating heart of many converters: the inductor. Buck, boost, grid filters, EMI chokes — inductors are everywhere, quietly deciding whether your converter behaves like a saint or a gremlin.

What is an inductor (quick, no fluff)

• Definition: An inductor is a two-terminal component that stores energy in a magnetic field when current flows through it. Its fundamental relation is (v(t) = L,\mathrm{d}i/\mathrm{d}t).
• Stored energy: (E = \tfrac{1}{2} L i^2).
• Magnetic parameter: (L = N^2 \dfrac{\mu A}{l}) (a useful engineering intuition: turns, permeability, cross-section up; length down).

Types, vibes, and when to pick them

Key practical behaviors (that will bite you if ignored)

• Saturation: Core materials have a max flux. Beyond that, L falls and current explodes. Use gapped cores for energy storage (boost inductors) to avoid sudden saturation.
• Core and copper losses: At high frequency, core loss (hysteresis + eddy currents) and skin/proximity effects in windings matter. Choose materials and litz wire accordingly.
• DC bias: Many ferrites lose inductance with DC current. The spec sheet is your friend.
• Parasitics: DCR (ohmic loss), stray capacitance, leakage inductance — these change resonances and EMI.

Inductors in converters — the bread and butter

Think of an inductor as both an energy bucket and a current smoother. In switching converters it does two crucial jobs:

1. Store/transfer energy (boost, buck-boost, SEPIC) — energy stored during one interval and released during another.
2. Limit di/dt and smooth current (buck, filters) — reduces ripple, protects switches, and shapes control dynamics.

Design formulae you will use in your sleep

• Inductor voltage law: v_L = L di/dt
• Energy stored: E = 1/2 * L * I^2
• Buck converter ripple (on-state):
  • Duty D = Vout/Vin
  • Ripple current: (\Delta I = \dfrac{(V_{in}-V_{out}) D}{L f_s})
• Boost converter (on-state ripple): (\Delta I = \dfrac{V_{in} D}{L f_s})
• LC cutoff: (f_c = \dfrac{1}{2\pi\sqrt{LC}})

Quick control note: In MPC and other predictive controllers, you explicitly predict the inductor current because it's a state with energy — so accurate L (including DC bias) matters or your predictions go to fantasy land.

Design example (you get to be the engineer)

Design target: Buck converter, Vin = 12 V, Vout = 5 V, Iout = 2 A, switching fs = 300 kHz, choose ripple = 30% of Iout (0.6 A).

• D = Vout/Vin = 5/12 ≈ 0.4167
• L = (Vin - Vout) * D / (ΔI * fs)
• L = 7 * 0.4167 / (0.6 * 300e3) ≈ 16.2 µH
• Peak current ≈ Iout + ΔI/2 = 2 + 0.3 = 2.3 A
• Energy at peak: E ≈ 0.5 * 16.2e-6 * 2.3^2 ≈ 43 µJ

Practical checks:

• Choose core with saturation current > 2.3 A (safety margin!).
• Check DCR to limit losses and temperature rise.
• Verify core loss at 300 kHz for expected flux density (ferrite might be OK; consider gapped powdered core if you need DC bias headroom).

Special flavors and applications

• Gapped power inductors: For energy-storage duties (boost converters). Gap stops magnetic coupling internally so the core won't saturate quickly.
• Chokes (EMI/Filter): Designed to present high impedance at switching harmonics. Differential-mode vs common-mode — differential chokes can be almost flux-cancelling for DC, common-mode do not.
• Current-sensing inductors (low-L): Used when you want a measurable di/dt behavior; sometimes intentionally low inductance used as a shunt alternative.
• Coupled inductors: Replace transformer in some topologies, reduce ripple, enable flux cancellation — useful in isolated PFC or SEPIC variants.

Control implications — bridging to what you learned in Control Strategies

• Inductors set the time constant L/R — slower dynamics require different controller bandwidths. If the inductor is large, your current loop must adapt.
• In Model Predictive Control, the inductor current is a predicted state. Prediction accuracy depends on L being stable under DC bias and temperature — otherwise MPC decisions become suboptimal.
• Grid-connected converters rely on L filters (often LCL) to meet harmonic injection limits. Proper inductor design and active damping strategies (control) reduce resonance — a cross-over with control strategies we discussed earlier.

"Designing inductors without thinking about the controller is like designing a car without brakes. Pretty until you hit 60 mph." — Your future exam grader

Practical checklist before you order the part

• Does L at operating DC current meet ripple spec? (check DC bias effect)
• Is saturation current > Ipeak with margin?
• Are core losses acceptable at switching frequency? (see datasheet loss curves)
• Is DCR acceptable for efficiency and temp rise?
• Does form factor and EMI behavior fit PCB layout needs? (toroids minimize stray flux)
• If using in grid-connected converter, check impedance vs frequency for resonance; coordinate with active damping in control design.

Final takeaway — the emotional truth

Inductors are simultaneously boring and dramatic: they quietly store energy, dictate dynamic response, and will sabotage your controller if their non-idealities are ignored. Treat them like the lead actor in the converter drama — they deserve design attention, respect for non-linear core physics, and a tight collaboration with your control strategy (especially MPC and grid converters).

Go pick a core, calculate ripple, and then pray to the datasheet gods. Or better: test with real parts.

Version note: builds directly on Magnetic Components > Transformers (you remember coupling vs storage) and Control Strategies (especially predictive control of inductor current and LCL-filter damping).