The resonant frequency of an LC circuit is dictated by the formula f = 1 / (2 * π * √(L * C)). If you need to design a standard 100 kHz RF tank or notch filter, you do not need to guess. By selecting a 100 pF capacitor and a 25.3 μH inductor in a parallel topology, you will hit 100.03 kHz. This guide breaks down exactly why you should choose a parallel topology over a series one for energy storage, how parasitic elements shift your target frequency, and the exact bench protocol to verify your design without accidentally detuning it with your test equipment.
Series vs. Parallel Topology: Which Node Configuration Wins?
Before picking components, you must define the topology. An LC circuit consists of two primary nodes and a ground reference. Let us label them: Node A (Signal Input/Source), Node B (The junction between the inductor and capacitor), and Node C (Ground/Return).
In a Series LC configuration, the inductor and capacitor are daisy-chained between Node A and Node C, with Node B acting as the midpoint. At resonance, the inductive reactance (XL) and capacitive reactance (XC) cancel each other out, resulting in near-zero impedance. This topology is ideal for bandpass filters where you want to pass the resonant frequency to ground or to the next stage with minimal voltage drop.
In a Parallel LC configuration (the "Tank" circuit), the inductor connects between Node B and Node C, and the capacitor also connects between Node B and Node C. Node A feeds into Node B, usually through a coupling resistor or a transistor collector/drain. At resonance, the impedance between Node B and Node C spikes to its maximum. The energy sloshes back and forth between the magnetic field of the inductor and the electric field of the capacitor.
If you are building an oscillator, a regenerative receiver, or an induction heater driver, the parallel LC tank is the mandatory choice. The high impedance at resonance provides the necessary voltage gain and phase shift for positive feedback. A series LC circuit would simply short your active device's output to ground at resonance, causing excessive current draw and potential thermal failure of your driving transistor. For a deep dive into tank circuit Q-factor dynamics, refer to the Electronics Tutorials guide on LC Oscillators.
Design Walkthrough: Picking Real Parts for 100 kHz
Let us design a parallel LC tank targeting exactly 100 kHz. We start by picking the capacitor, as standard capacitor values are more rigidly spaced (E12/E24 series) than inductors.
- Select C: We choose 100 pF. Going lower (e.g., 10 pF) makes the circuit overly susceptible to stray breadboard capacitance. Going higher (e.g., 10 nF) requires physically large inductors with high DC resistance (DCR), which kills the Q-factor.
- Calculate L: Rearranging the formula to L = 1 / ((2 * π * f)² * C).
L = 1 / ((6.283 * 100,000)² * 100e-12) = 25.33 μH. - Select the Dielectric: You must use a C0G/NP0 ceramic dielectric. If you use an X7R or Y5V capacitor, the capacitance will drop drastically under DC bias or shift with temperature, detuning your frequency of the LC circuit mid-operation.
Concrete Component Pick:
- Capacitor: Murata GRM1555C1H101JA01D (100 pF, 50V, C0G, 0402 SMD) or the through-hole equivalent Kemet C315C101J1G5TA (100 pF, 100V, C0G). Cost: ~$0.10.
- Inductor: Bourns 78F250K-RC (25 μH, 10% tolerance, axial leaded). Cost: ~$0.45. This is a molded inductor, which shields the magnetic field and prevents it from coupling with nearby breadboard traces.
Behavior Matrix: What Happens When Components Shift
Real-world components are not ideal. Parasitics will alter the theoretical frequency of your LC circuit. Here is how the topology behaves when non-ideal factors are introduced.
| Variable Shift | Effect on Resonant Frequency | Effect on Tank Q-Factor | Physical Cause on the Bench |
|---|---|---|---|
| Inductance (L) Increases | Decreases | Decreases (usually) | Core material permeability rises with temperature; physical deformation of coil. |
| Capacitance (C) Increases | Decreases | Minimal direct effect | Stray capacitance from scope probes, breadboard rails, or X7R dielectric DC bias. |
| Inductor ESR Increases | No direct shift | Severe Decrease | Skin effect at high frequencies; using a ferrite core with high hysteresis loss at 100 kHz. |
| Capacitor ESL Increases | Increases (creates secondary resonance) | Degraded at high freq | Using long lead lengths on through-hole capacitors instead of SMD or trimming leads flush. |
Failure Modes: What Breaks at the Extremes?
When troubleshooting a dead board, you need to know how the circuit behaves when an element catastrophically fails. A multimeter in resistance mode will reveal these extremes.
- Capacitor Shorts (Node B to Node C reads 0 Ω): The resonant frequency drops to 0 Hz. In a parallel tank driven by a transistor, this effectively ties the collector/drain directly to ground. The driving transistor will likely overheat and fail due to unimpeded DC current flow.
- Capacitor Opens (Node B to Node C reads infinite Ω for the C branch): The LC resonance is destroyed. The circuit becomes purely inductive. The phase shift required for oscillation will fail to materialize, and the oscillator will simply output a flat DC rail voltage.
- Inductor Shorts (Winding insulation fails): The inductance drops near zero, pushing the theoretical frequency toward infinity. Practically, the circuit becomes purely capacitive, and the low DC resistance of the shorted inductor will draw massive current from the bias network, potentially burning out the biasing resistor.
- Inductor Opens (Wire breaks inside the mold): The DC path to ground is severed. Node B will float up to the supply rail voltage. No AC signal will be generated or filtered.
Step-by-Step Breadboard Testing Protocol
Testing a high-Q parallel LC tank on a solderless breadboard introduces significant measurement errors if done incorrectly. The Analog Devices MT-042 Tutorial highlights how test equipment loading can ruin resonance measurements. Follow this exact sequence:
Never connect a function generator's 50-ohm output directly across a parallel LC tank. The 50-ohm source impedance will dominate the circuit, crushing the Q-factor and flattening the resonance peak so much that you will not see it on your scope.
- Insert Components: Plug the Bourns 25 μH inductor and Kemet 100 pF capacitor into the same breadboard rows to form Node B. Connect the other legs to the ground rail (Node C).
- Isolate the Source: Connect your function generator's output to Node A. Insert a 1 kΩ series resistor between Node A and Node B. This converts the low-impedance voltage source into a high-impedance current source, preserving the tank's Q-factor.
- Configure the Scope: Attach a 10x oscilloscope probe to Node B. Bench Insight: A standard 10x probe adds ~12 pF of capacitance. This will artificially increase your 100 pF tank to 112 pF, dropping your measured frequency to ~94 kHz. If you need exact verification, use an active FET probe (< 2 pF) or mathematically subtract the probe capacitance from your calculations.
- Sweep the Frequency: Set the function generator to a 1Vpp sine wave. Start a logarithmic sweep from 10 kHz to 1 MHz.
- Identify the Peak: Watch the oscilloscope amplitude. The voltage at Node B will peak sharply at resonance. If using the 1 kΩ isolation resistor and a 10x probe, expect the peak to land between 94 kHz and 98 kHz due to probe loading and the 10% tolerance of the molded inductor.
Decision Tree: Finalizing Your Component Pick
Use this decision matrix to lock in your final Bill of Materials (BOM) based on your specific application constraints. Do not default to 'it depends'—pick the row that matches your primary constraint and order the parts.
| Primary Constraint | If your priority is... | Then select this Topology | Concrete Component Pick (100 kHz Target) |
|---|---|---|---|
| Maximum Q-Factor (RF Oscillators) | Minimal ESR, tight frequency stability, low microphonics. | Parallel Tank | C: Murata C0G 100pF (GRM1555C1H101JA01D) L: Coilcraft 1606-A (25 μH, powdered iron, high Q) |
| High Current Handling (Induction Heating) | Surviving 10A+ circulating currents without core saturation. | Parallel Tank | C: WIMA Snubber 100nF (film, high pulse) L: Custom wound on Amidon T-106-2 powdered iron toroid |
| Cost & Space (Simple IF Filters) | Lowest BOM cost, acceptable 10% frequency drift. | Series or Parallel | C: Yageo X7R 100pF (0603) L: Bourns 78F250K-RC (25 μH axial molded) |
For a standard hobbyist or prototyping environment targeting the frequency of an LC circuit at 100 kHz, the Bourns 78F250K-RC paired with a C0G 100 pF capacitor is the definitive, default choice. It provides shielded magnetic properties to prevent breadboard crosstalk, survives standard 5V logic-level RF driving stages, and keeps your BOM cost under a dollar while maintaining a predictable resonance peak.
