An air inductor uses air—or a non-magnetic structural form like ceramic, plastic, or brass—as its core medium. Because it lacks a magnetic core, it offers infinite saturation current and zero hysteresis loss, making it the undisputed choice for high-Q (quality factor) RF circuits above 10 MHz. The tradeoff is low permeability ($\mu_r \approx 1$), meaning you need significantly more turns of wire to achieve the same inductance as a ferrite or iron-powder core. If you are designing matching networks, VCO tanks, or HF antennas, understanding the physical realities of air-core coils is the difference between a circuit that works on paper and one that works on the bench.
Inductor Core Types: Where Air Wins (and Loses)
Choosing the right core material is about balancing inductance density against frequency limitations. Magnetic cores multiply inductance but introduce losses that scale non-linearly with frequency. Here is how air stacks up against common magnetic alternatives when deciding which type for which job.
| Core Type | Construction | Typical Tolerance | Tempco (ppm/°C) | Typical Use Case |
|---|---|---|---|---|
| Air / Ceramic | Enamelled wire wound on non-magnetic form or self-supporting | ±2% to ±5% | +50 to +150 (copper dependent) | RF tank circuits, VCOs, >10 MHz matching networks, UHF antennas |
| Ferrite (NiZn) | Nickel-zinc ceramic bobbin or toroid | ±5% to ±20% | +100 to +500 | EMI suppression beads, broadband RF transformers (1 MHz - 100 MHz) |
| Ferrite (MnZn) | Manganese-zinc ceramic bobbin | ±10% to ±30% | +500 to +3000 | Switch-mode power supply (SMPS) chokes, low-frequency filters (<1 MHz) |
| Iron Powder | Insulated iron particles pressed into a toroid or rod | ±10% to ±15% | +20 to +150 | High-power RF amplifiers, HF impedance matching (1 MHz - 30 MHz) |
| Toroidal (Tape) | Silicon steel or amorphous metal tape wound on a spool | ±15% to ±25% | Variable | 50/60 Hz line filters, high-current common-mode chokes |
Decoding the Markings and Specifications
Reading inductor markings is notoriously frustrating because the industry lacks a single, universally enforced standard like the EIA resistor color code. What the markings mean depends heavily on the physical form factor.
SMD RF Chip Inductors
For surface-mount RF inductors (like the Coilcraft 0603CS series), you will typically see a three-digit code. The first two digits are the significant figures, and the third digit is the multiplier in nanohenries (nH).
- 101 = 10 × 10¹ = 100 nH
- 471 = 47 × 10¹ = 470 nH
- 100 = 10 × 10⁰ = 10 nH (Note: In power inductors, '100' often means 10 µH. Always check the datasheet if the physical size suggests a power choke rather than an RF part).
Axial Leaded RF Chokes
Molded axial inductors often use four color bands, identical to the resistor color code, but the resulting value is read in microhenries (µH). A brown-black-brown-gold band translates to 1-0-1 (100 µH) with a ±5% tolerance. For values under 10 µH, a decimal point is implied; a black-gold-gold band means 0.0 µH, which is practically a jumper wire used for its specific parasitic inductance.
Bench Scenario: The 13.56 MHz RFID Matching Network Failure
To understand why core material matters, let us walk through a real-world bench failure involving an NFC/RFID reader operating at 13.56 MHz.
The Setup: I was designing an L-match network to transform a 50-ohm transceiver output to a low-impedance PCB trace antenna. The target inductance was 470 nH. To save board space, I initially selected a shielded, ferrite-core SMD power inductor rated for 1.5A DC.
The Numbers: At 13.56 MHz, a 470 nH inductor has a theoretical reactance ($X_L = 2\pi f L$) of roughly 40 Ω. For a sharp resonance peak and maximum magnetic field radiation, the inductor needed a Quality Factor (Q) of at least 60.
The Outcome: The reader's range plummeted from an expected 50 mm down to 8 mm. The transceiver IC was running hot, and the antenna matching network felt warm to the touch.
What Went Wrong: The ferrite core in the SMD part was optimized for DC-DC switching regulators (typically 100 kHz to 2 MHz), not 13.56 MHz RF. At 13.56 MHz, the ferrite introduced massive hysteresis and eddy current losses. When I put the board on an impedance analyzer, the measured Q factor of the ferrite inductor was just 12.
The Fix: I desoldered the SMD part and wound a custom air-core coil using 7 turns of 20 AWG enameled copper wire on a 5mm ceramic form. The Q jumped to 82, and the read range stabilized at 55 mm. You can read more about high-frequency core losses in the Coilcraft inductor design library, which provides excellent empirical data on core material limits.
Failure Modes and Visual Diagnostics
Air inductors do not suffer from magnetic saturation or core cracking, but they are highly susceptible to mechanical and parasitic failures. Here is how to diagnose them on the bench.
| Failure Mode | Visual Symptom | Electrical Consequence |
|---|---|---|
| Mechanical Deformation | Coil looks squeezed, bent, or has uneven turn spacing. | Inductance shifts by 10-30%. In a tank circuit, this detunes the resonant frequency entirely. |
| Microphonics | Loose wire loops; lack of rigid potting, dope, or conformal coating. | Vibrations change turn spacing dynamically, causing frequency modulation (FM noise) in oscillators. |
| Parasitic Self-Resonance | Multi-layer overlapping windings or excessively tight turn spacing. | Inter-winding capacitance creates a parallel resonant circuit. Above the Self-Resonant Frequency (SRF), the inductor behaves like a capacitor. |
| Skin Effect Overheating | Thick solid wire used at VHF/UHF frequencies; wire discoloration at the center. | Current flows only on the outer skin of the wire, drastically increasing AC resistance and lowering Q. |
For a deeper theoretical breakdown of parasitic capacitance and self-resonance in passive components, the All About Circuits textbook on AC theory provides excellent foundational models.
Safe Substitution: Winding Your Own Air Inductor
When you are prototyping and do not have the exact 470 nH ±5% ceramic chip inductor in your bins, you can safely substitute it by winding your own air-core coil. Because air has a permeability of exactly 1, the math is highly predictable.
The Bench Formula
For single-layer solenoid air inductors, use Wheeler's Approximation. It is accurate to within 1% for coils where the length is greater than 0.8 times the radius.
$L (\mu H) = \frac{r^2 \times N^2}{9r + 10l}$
- L = Inductance in microhenries (µH)
- r = Radius of the coil in inches
- N = Number of turns
- l = Length of the winding in inches
Step-by-Step Winding Procedure
- Select the Wire: For frequencies below 30 MHz, standard enameled copper wire (magnet wire) is fine. For VHF/UHF (>50 MHz), use Litz wire or silver-plated copper to mitigate skin effect losses.
- Choose the Form: Use a non-magnetic mandrel. Ceramic, Teflon, or polycarbonate are ideal. Avoid standard PVC as it has high dielectric losses at RF. If the inductance is small (<100 nH), wind it as a self-supporting 'pigtail' without a form.
- Wind and Space: Wind the turns tightly for maximum inductance density, but if you need to push the Self-Resonant Frequency (SRF) higher, space the turns apart by at least one wire diameter to reduce inter-winding capacitance.
- Lock it Down: Once you hit the target value on your LCR meter, apply a drop of low-loss RF dope, cyanoacrylate (super glue), or clear nail polish to freeze the turns in place. This prevents microphonics and mechanical drift.
- Verify with an LCR Meter: Always measure the substituted part at the actual operating frequency if your meter supports it. Inductance measured at 1 kHz can differ from 13.56 MHz due to parasitic effects.
Substituting an air inductor is one of the few times in electronics where 'homebrew' parts frequently outperform mass-manufactured components, provided you respect the geometry and secure the windings against physical vibration.
