The Verdict: When to Wire Inductors in Series vs Parallel
When deciding between wiring inductors in series vs parallel, series is the clear winner for 95% of applications, specifically when you need to increase total inductance, boost high-frequency impedance for EMI filtering, or hack together a high-L value using cheaper, off-the-shelf parts. Parallel wiring wins only in extreme high-current edge cases where you must share massive RMS ripple current across multiple smaller footprints because a single inductor with the required current rating would physically not fit on the PCB or would suffer from severe core saturation.
Choose Series When:
- You need a higher inductance value than what is available in a single standard package.
- You are building a multi-stage LC low-pass filter and need to maximize impedance to block high-frequency switching noise.
- You want to use two smaller, readily available shielded SMT inductors to save BOM costs and avoid long lead times on custom or massive toroidal parts.
Choose Parallel When:
- You are forced to split extreme RMS current (e.g., >30A) across two smaller inductors because a single inductor with that current rating and required inductance would saturate or exceed your PCB height restrictions.
- You are designing a multiphase interleaved buck converter where each phase requires its own dedicated inductor tied to a common output node (though technically these are parallel phases, not a single parallel inductor pair).
The Single Physical Difference Driving the Math
The entire behavioral divergence between series and parallel inductors comes down to magnetic flux linkage and current division.
In a series configuration, the exact same current flows sequentially through both components. Because inductance is fundamentally a measure of how much magnetic flux is generated per ampere of current ($L = N\Phi / I$), forcing the same current through two discrete magnetic cores causes their individual flux fields to stack. Assuming they are physically spaced apart to prevent mutual coupling, their inductances add linearly: $L_{total} = L_1 + L_2$. Think of it like water flowing through a single pipe that passes through two separate restriction valves; the total restriction (inductance) is the sum of both.
In a parallel configuration, the voltage potential across both inductors is identical, but the current divides between them. The current takes the path of least impedance, meaning it splits inversely proportional to their inductance values (and critically, their DC resistance). The total equivalent inductance drops, following the product-over-sum formula for two identical values: $L_{total} = (L_1 \times L_2) / (L_1 + L_2)$. If you parallel two 10µH inductors, you get 5µH. The magnetic fields are generated independently by fractional currents, reducing the overall opposition to changes in current.
For a deeper mathematical breakdown of these foundational rules, the All About Circuits textbook chapter on series and parallel inductors provides excellent baseline formulas.
Head-to-Head Comparison: Series vs Parallel Inductors
Here is how the two configurations stack up across the criteria that actually matter on the bench and in production.
| Criteria | Inductors in Series | Inductors in Parallel |
|---|---|---|
| Total Inductance | Adds up ($L_1 + L_2$). Increases total L. | Drops ($L_1 \times L_2 / (L_1 + L_2)$). Decreases total L. |
| Total DCR (DC Resistance) | Adds up ($R_1 + R_2$). Increases $I^2R$ conduction losses. | Divides (if matched). Lowers overall DCR and conduction losses. |
| Saturation Current ($I_{sat}$) | Limited by the weakest inductor in the chain. | Complex; depends on exact current division. Can theoretically double if perfectly matched. |
| High-Freq Impedance | High. Excellent for choking EMI and switching noise. | Low. Acts as a bypass for high-frequency AC ripple. |
| PCB Footprint Efficiency | Poor. Takes up 2x the board space for a given L value. | Moderate. Used only when single large inductors cannot fit. |
Where These Configurations Are NOT Interchangeable
You cannot simply swap a series inductor pair for a parallel pair and adjust the values to compensate. The failure modes are entirely different.
The Parallel DCR Mismatch Trap
The most common bench failure when wiring inductors in parallel is DCR thermal runaway. Inductors are not ideal components; they have parasitic DC resistance (DCR). If you parallel two 10µH inductors, but Inductor A has a DCR of 4mΩ and Inductor B has a DCR of 8mΩ (due to manufacturing tolerances or temperature gradients on the PCB), the DC load current will not split 50/50. It will split based on the resistance ratio. Inductor A will hog 66% of the DC current. It will heat up faster, which increases its copper DCR further, potentially leading to core saturation. Once Inductor A saturates, its inductance drops to near zero, dumping all the AC ripple current into Inductor B, which then overheats and fails. According to the Coilcraft power inductor design guide, paralleling inductors requires obsessive attention to DCR matching and thermal symmetry, which is rarely achievable on standard FR4 PCBs.
Resonant Tanks and LLC Converters
In resonant topologies like LLC converters, the precise inductance value dictates the switching frequency and zero-voltage-switching (ZVS) boundaries. Wiring inductors in series introduces parasitic inter-winding capacitance that shifts the self-resonant frequency (SRF) unpredictably. Parallel inductors introduce mutual coupling risks if placed close together, altering the tank dynamics. In these circuits, a single, precision-gapped custom inductor is mandatory.
Cost, Availability, and the 'Two-Cheaper-Parts' Hack
One of the most practical reasons to wire inductors in series is BOM (Bill of Materials) optimization. High-inductance, high-current power inductors require massive ferrite cores and thick copper windings. They are expensive, heavy, and frequently suffer from 12-to-20-week lead times.
The Numeric Example: Suppose your 48V-to-12V buck converter needs 47µH at 15A RMS. A single premium part like the Bourns SRP1265A-470M might cost $3.80 in low volumes and show a 14-week lead time at your distributor. However, you can use two Coilcraft XGL6060-230 (23µH, 16A $I_{rms}$, shielded) inductors wired in series.
- Total Inductance: 23µH + 23µH = 46µH (close enough for the control loop to compensate).
- Current Handling: 16A (limited by the single part rating, which exceeds your 15A requirement).
- Total Cost: ~$1.15 each = $2.30 total.
- Availability: Usually in stock at Digi-Key or Mouser.
This series hack saves you $1.50 per board and eliminates supply chain risk. Just ensure you physically space them apart or use shielded cores (like the XGL series) to prevent mutual inductance from altering your target value. For more on component selection and magnetics, Electronics Tutorials offers solid primers on core materials and shielding.
Frequently Asked Questions
Do inductors in series add up exactly like resistors?
Mathematically, yes ($L_{total} = L_1 + L_2 + ...$). Physically, no. Resistors do not interact with each other via invisible fields. Inductors generate magnetic flux. If you place two unshielded inductors in series too close together on a PCB, their magnetic fields will couple. This mutual inductance ($M$) means the total inductance becomes $L_{total} = L_1 + L_2 \pm 2M$, depending on their physical orientation. To ensure they add up like resistors, you must use magnetically shielded cores or space them far apart and rotate them 90 degrees to each other.
What happens to mutual inductance when wiring inductors in series vs parallel?
Mutual inductance affects both configurations, but the consequences differ. In series, mutual coupling adds to or subtracts from the total inductance, potentially pushing your LC filter cutoff frequency out of spec. In parallel, mutual coupling can cause severe circulating currents between the two components, especially during high $di/dt$ switching transients, leading to localized heating and EMI radiation. Parallel inductors must always be shielded or placed on opposite sides of the PCB.
Can I put two different value inductors in parallel?
You can, but it is highly discouraged for power circuits. In a parallel setup, AC ripple current divides inversely to the inductance value. If you parallel a 10µH inductor with a 40µH inductor, the 10µH inductor will absorb 80% of the high-frequency AC ripple current. It will likely exceed its ripple current rating, overheat, and saturate long before the 40µH inductor is stressed. Always use identical part numbers when paralleling.
Does wiring inductors in series increase the current rating?
No. This is a dangerous misconception. The current rating of a series string is strictly limited by the weakest link—specifically, the component with the lowest saturation current ($I_{sat}$) or RMS thermal current rating ($I_{rms}$). If you put a 20A inductor in series with a 5A inductor, the entire chain is limited to 5A. If you exceed 5A, the smaller inductor's core will saturate, its inductance will collapse, and it will likely overheat and fail, taking the whole circuit down with it.
