What is a Coupled Inductor and Why Use It Over Discrete Chokes?

A coupled inductor is a single magnetic component containing two or more electrically isolated windings wrapped around a shared magnetic core. Unlike a standard transformer, which is designed to transfer energy from primary to secondary with minimal energy storage, a coupled inductor is often designed to store energy in its core gap (as in flyback topologies) or to pass energy through while maintaining matched inductance (as in SEPIC, Cuk, and Zeta converters).

The primary engineering advantage of using a coupled inductor over two discrete chokes is the reduction of ripple current. In a SEPIC (Single-Ended Primary-Inductor Converter) topology, tight magnetic coupling between the two windings allows the ripple currents to partially cancel each other out. This ripple steering effect allows you to use smaller output capacitors and achieve lower electromagnetic interference (EMI). Furthermore, it halves the component count, reduces PCB footprint, and guarantees that the inductance values of both windings track perfectly across temperature variations.

Warning: Never confuse a coupled inductor with a forward-mode transformer in a flyback design. While they look identical on a schematic, a flyback 'transformer' is technically a coupled inductor that requires a physical air gap in the core to store energy. Using a standard ungapped transformer in a flyback circuit will result in immediate core saturation and catastrophic MOSFET failure.

Coupled Inductor Selection Matrix: Core Types and Real-World Specs

Selecting the right physical construction dictates your thermal performance, EMI profile, and saturation limits. Below is a data-dense comparison of the four primary coupled inductor constructions available for modern power supply designs.

Core Construction Example Part Family Inductance Range Tolerance / Tempco Isat / Irms Limits Typical Topology
Shielded Drum Coilcraft MSD1260 4.7 µH to 1 mH ±20% / N/A (Ferrite) Up to 9A Isat SEPIC, Flyback
Unshielded Drum Bourns SRN6045TA Dual 10 µH to 470 µH ±20% / N/A Up to 3A Isat Low-cost SEPIC
Toroidal Würth WE-FLY 10 µH to 10 mH ±15% / Low leakage Up to 5A Irms Common-mode, Cuk
Planar / PCB TDK SPM6530T Dual 1.0 µH to 47 µH ±20% / High freq Up to 12A Isat High-freq Zeta/SEPIC

Which Type for Which Job?

  • Shielded Drum Core: Choose this when EMI is a concern and you are operating in a noise-sensitive environment (e.g., medical or audio equipment). The magnetic shield prevents flux fringing from inducing noise into nearby traces.
  • Unshielded Drum Core: Best for cost-sensitive, space-constrained consumer electronics where EMI can be managed at the system level. Beware of flux fringing causing eddy current heating in nearby copper pours.
  • Toroidal: Ideal when you need an exceptionally high coupling coefficient (k > 0.99) and minimal leakage inductance. Often used in high-efficiency Cuk converters where leakage inductance directly degrades performance.
  • Planar/Multilayer: Select for ultra-high switching frequencies (2 MHz to 5 MHz) using GaN or SiC switches. The low profile and distributed capacitance minimize high-frequency ringing.

Decoding Part Markings and the Dot Convention

Reading the markings on a coupled inductor is critical for both identifying the value and ensuring correct phase orientation on the PCB.

The Inductance Code

For SMD coupled inductors, manufacturers typically use a three-digit code based on microhenries (µH), which differs from the resistor coding system. The first two digits are the significant figures, and the third digit is the multiplier (number of zeros).

  • 100 = 10 × 10^0 = 10 µH
  • 471 = 47 × 10^1 = 470 µH
  • 152 = 15 × 10^2 = 1500 µH (1.5 mH)

Larger through-hole or heavily shielded parts (like the Coilcraft MSD series) often bypass the code and print the value directly (e.g., '100µH') alongside a date or lot code. Always verify with an LCR meter if the marking is ambiguous.

The Dot Convention and Phasing

The 'dot' on a coupled inductor schematic symbol indicates the relative winding direction. Physically, this is marked on the component body with a painted dot, a beveled edge, or a specific pin 1 indicator (often a chamfered corner on the plastic mold).

Bench Rule: In a SEPIC converter, the dotted ends of both windings must connect to the input voltage and the switch node, respectively. If you reverse the phasing (connect a dot to ground), the mutual inductance will fight the self-inductance during the switch-on time. The effective inductance drops to near zero, causing massive current spikes that will instantly destroy your switching MOSFET.

Safe Substitution: What to Do When the Exact Part is Obsolete

Supply chain disruptions frequently force engineers to substitute magnetics. You cannot simply swap a coupled inductor based on matching the inductance value alone. Follow this hierarchy to safely substitute a part:

  1. Match Inductance (L) and DCR: The nominal inductance of each winding must be within 10% of the original. The DC Resistance (DCR) should be equal to or lower than the original to prevent thermal runaway.
  2. Verify Current Ratings (Isat and Irms): Isat (saturation current) is the point where inductance drops by 20-30%. Irms is the thermal limit where the part rises 40°C. Your substitute must meet or exceed both limits for your specific topology's peak and average currents.
  3. Check the Coupling Coefficient (k) and Leakage Inductance: This is where most substitutions fail. Calculate the coupling coefficient using the formula: k = M / √(L1 × L2). If your original design relied on tight coupling (k > 0.98) to minimize voltage spikes, substituting a loosely coupled part (k = 0.92) will increase leakage inductance. In a flyback converter, higher leakage inductance requires a heavier RCD snubber, reducing efficiency. In a SEPIC, it causes high-frequency ringing on the AC coupling capacitor.
  4. Confirm Footprint and Pinout: Even if the electrical specs match, a mirrored pinout will force you to cross traces on the PCB, which can introduce parasitic loop inductance and ruin your EMI performance.

Failure Modes and Bench-Level Diagnostics

Coupled inductors are generally reliable, but they are subject to severe electrical and thermal stresses in switching power supplies. When a power supply fails, use this diagnostic matrix to isolate the inductor fault.

Failure Mode Visual / Olfactory Symptoms Root Cause Bench Diagnostic Test
Core Saturation No initial visual signs; secondary symptom is a shattered switching MOSFET or blown input fuse. Peak current exceeded Isat limit, often due to a shorted output load or incorrect air gap. Measure inductance with an LCR meter at 100 kHz / 1 Vrms. If L reads >20% below nominal at zero bias, the core may be permanently magnetized or cracked.
Inter-Winding Short Burnt enamel smell, blackened core surface, or melted plastic bobbin near the pin terminals. High dv/dt voltage spikes broke down the insulation between the primary and secondary windings. Use a multimeter in continuity mode between Pin 1 (Winding A) and Pin 3 (Winding B). Any reading below infinite resistance (OL) confirms a short.
Intra-Winding Short Component runs excessively hot (>100°C) under light load; converter output voltage sags. Insulation failure between adjacent turns of the same winding, effectively reducing the turn count. Measure DCR. A significantly lower DCR than the datasheet spec, combined with a lower measured inductance, indicates shorted turns.
Winding Open No output voltage; no physical damage visible to the naked eye. Mechanical stress during PCB assembly (bending the board) or thermal cycling cracked the wire at the termination point. Multimeter continuity test across pins 1-2 and 3-4. An open circuit (OL) reading confirms a broken termination.

For advanced diagnostics, measuring leakage inductance is a powerful troubleshooting step. To measure leakage inductance on your bench, short-circuit the secondary winding (pins 3 and 4) using a thick piece of bare copper wire. Then, measure the inductance across the primary winding (pins 1 and 2) using your LCR meter set to 100 kHz. The value you read is the primary leakage inductance. If this value has drifted significantly from your initial prototype baseline, the core may have suffered mechanical micro-fractures from thermal expansion, altering the magnetic coupling path.

For further reading on optimizing magnetic components in non-isolated topologies, refer to the Texas Instruments Power Design Support documentation, which provides excellent baseline calculations for coupled inductor ripple current.