A resonant frequency circuit exploits the continuous energy exchange between an inductor’s magnetic field and a capacitor’s electric field. At the resonant frequency ($f_r$), the inductive reactance ($X_L$) and capacitive reactance ($X_C$) are equal in magnitude but opposite in phase, effectively canceling each other out. The governing equation is $f_r = 1 / (2\pi\sqrt{LC})$. While the math is straightforward, translating this into a physical circuit on a workbench requires navigating parasitic capacitance, component tolerances, and dielectric absorption. This guide walks through the physical design, component selection, and bench verification of a 100 kHz parallel resonant tank—a foundational topology for RF filters, wireless power transfer, and oscillator networks.
Parallel vs. Series RLC: Why Choose the Parallel Tank Topology?
Before selecting components, you must choose between a series or parallel RLC configuration. The decision hinges entirely on the impedance behavior you need at resonance and how the circuit interfaces with your source and load.
In a series resonant circuit, the components are daisy-chained between the source and ground. At $f_r$, the reactances cancel, leaving only the parasitic DC resistance (ESR) of the components. This results in minimum impedance and maximum current flow. Series topologies are ideal for notch filters, induction heating inverters, and applications where you need to pass a specific frequency while blocking others.
In a parallel resonant circuit (often called a tank circuit), the inductor and capacitor are wired in parallel with each other, and the combination is placed in shunt across the signal path or driven by a current source. At $f_r$, the circulating current between the L and C components is maximized, but the current drawn from the external source drops to a minimum. This results in maximum impedance. We choose the parallel topology when we need a bandpass filter, an LC oscillator, or an impedance-matching network for antennas, because the high impedance at resonance allows the signal voltage to peak dramatically without loading down the driving source.
Designing a 100 kHz Resonant Frequency Circuit: Real Component Selection
Let’s design a parallel tank circuit tuned to 100 kHz, a common intermediate frequency (IF) and RFID operating point. We will define three nodes: Node A (AC Source Input), Node B (Top of the Tank / Output Measurement), and GND (Bottom of the Tank / Circuit Common). A feed resistor ($R_{feed}$) connects Node A to Node B, while the inductor ($L$) and capacitor ($C$) connect between Node B and GND.
To hit 100 kHz, we need to select standard, readily available component values. Let’s fix the inductor at 250 µH. Rearranging the resonance formula to solve for capacitance yields approximately 10.13 nF. We will use a standard 10 nF capacitor.
| Component | Node Map | Target Value | Real-World Part / Spec | Critical Parameter |
|---|---|---|---|---|
| Inductor (L1) | Node B to GND | 250 µH | Bourns 78FR250K-RC (Radial) | SRF must be > 1 MHz; Q > 40 |
| Capacitor (C1) | Node B to GND | 10 nF (103) | Kemet C315C103J1G5TA | Must be C0G/NP0 dielectric, 50V |
| Feed Resistor (R1) | Node A to Node B | 10 kΩ | Yageo CFR-25JR-52-10K | Sets the Q-factor and bandwidth |
| Source (V1) | Node A to GND | 1 Vpp Sine | Siglent SDG1032X (50Ω out) | Sweep 10 kHz to 500 kHz |
The Dielectric Trap: Why C0G/NP0 is Mandatory
The most common mistake hobbyists make when building high-Q resonant circuits is grabbing a generic X7R or Y5V ceramic capacitor from their assortment kit. X7R dielectrics exhibit a severe voltage coefficient. At the high AC voltages generated across the tank at resonance, an X7R capacitor's actual capacitance can drop by 15% to 30%. This shifts your 100 kHz resonant peak up to 115 kHz or higher, completely detuning the circuit. You must specify C0G (also known as NP0) dielectrics for the tank capacitor. C0G capacitors have a near-zero temperature and voltage coefficient, ensuring the 10 nF value remains stable regardless of the circulating tank voltage.
Component Drift and Failure Modes: What Breaks at the Extremes?
On the bench, components drift due to temperature, aging, and physical stress. Understanding how the circuit behaves when a parameter shifts—or when a component fails catastrophically—is essential for troubleshooting.
Parameter Drift Behavior Matrix
| Element Change | Effect on Resonant Freq ($f_r$) | Effect on Q-Factor | Physical Cause on the Bench |
|---|---|---|---|
| Inductance (L) Increases | Decreases | Minor decrease | Core saturation, or moving a ferromagnetic tool near the inductor. |
| Capacitance (C) Increases | Decreases | Unchanged | Adding scope probe capacitance (10-15pF) to Node B. |
| Feed Resistance (R) Increases | Unchanged | Increases (sharper peak) | Using a higher value feed resistor to reduce source loading. |
| Inductor ESR Increases | Unchanged | Decreases (broader peak) | Inductor overheating, increasing copper winding resistance. |
Catastrophic Failure Modes (Open vs. Short)
When debugging a dead board, you need to know what a multimeter will read when a component fails.
- Capacitor Opens: The tank is broken. Node B simply sees the inductor to ground. The circuit loses resonance entirely and behaves as a simple first-order low-pass RL filter. The voltage peak at 100 kHz vanishes.
- Capacitor Shorts: Node B is hard-shorted to GND. The 10 kΩ feed resistor ($R_{feed}$) will dissipate all the source power ($P = V^2/R$). If driven by a low-impedance source without a feed resistor, the driver IC will likely overcurrent and shut down or burn out.
- Inductor Opens: Similar to the capacitor opening, the tank is broken. Node B sees only the capacitor to ground. The circuit becomes a first-order high-pass RC filter. No resonance peak will be observed.
- Inductor Shorts (Rare): Inductors usually fail open, but if the enamel insulation melts and windings short, the inductance drops drastically. The resonant frequency will spike into the MHz range, and the Q-factor will collapse due to the extremely low parallel impedance.
Breadboard Verification: Step-by-Step Testing with a Scope
Theory and simulation only get you so far. Parasitic capacitance from the breadboard traces (typically 2pF to 5pF per node) and the oscilloscope probe itself will slightly detune the circuit. Here is how to empirically verify your parallel resonant circuit on the bench.
Required Bench Equipment
- Function Generator (e.g., Siglent SDG1032X) with a 50Ω output.
- Digital Oscilloscope (e.g., Rigol DS1054Z) with a 10X passive probe.
- Solderless breadboard and short jumper wires.
Testing Procedure
- Build the Tank: Insert the 250 µH inductor and 10 nF C0G capacitor into the breadboard so both share a common ground rail. Connect the 10 kΩ feed resistor from the signal input rail to the top of the tank (Node B).
- Connect the Source: Attach the function generator's BNC-to-alligator cable. Connect the center conductor to Node A (the free leg of the feed resistor) and the shield to the ground rail.
- Configure the Generator: Set the function generator to output a 1 Vpp sine wave. Do not enable the 50Ω output impedance matching on the generator menu unless you are terminating the line; leave it in High-Z mode so the display reads the actual open-circuit voltage.
- Probe Node B: Attach the 10X oscilloscope probe to Node B. Crucial: Ensure the probe is set to 10X and the scope channel is configured to 10X. A 1X probe will add ~100pF of capacitance, dragging your 100 kHz resonance down to roughly 90 kHz and heavily damping the Q-factor.
- Execute the Sweep: Slowly sweep the function generator frequency from 10 kHz up to 300 kHz. Watch the peak-to-peak voltage on the oscilloscope.
- Identify the Peak: You will observe the voltage at Node B rise, peak sharply, and then fall. The frequency at the maximum voltage amplitude is your empirical $f_r$. Note this value (it should read between 98 kHz and 102 kHz, accounting for 5% component tolerances).
- Measure the Bandwidth: To calculate the real-world Q-factor, find the two frequencies on either side of the peak where the voltage drops to 70.7% (-3dB) of the peak value. Subtract the lower frequency from the higher frequency to get the Bandwidth (BW). Calculate $Q = f_r / BW$. With a 10 kΩ feed resistor and low-ESR components, you should measure a Q between 50 and 65.
By following this exact topology and respecting the dielectric requirements of the tank capacitor, you move beyond abstract textbook formulas and build a resonant network that behaves predictably under real-world bench conditions.
