The cutoff frequency of a first-order passive RC low pass filter is calculated using the formula fc = 1 / (2πRC). At this exact frequency, the output signal is attenuated by -3dB, meaning the output voltage drops to 70.7% of the input voltage. This formula is the absolute baseline for designing audio crossovers, smoothing PWM signals into analog DC, and knocking out high-frequency EMI noise before it hits a sensitive ADC pin.
Below, we break down the derivation, track units through real-world bench problems, and provide a concrete decision path for picking standard E-series component values.
Derivation and Symbol Definitions
The low pass filter formula is not an arbitrary rule; it is derived from the voltage divider principle applied to a resistor and a capacitor in series. The output is taken across the capacitor. The cutoff frequency (the -3dB point) occurs precisely when the magnitude of the capacitive reactance (XC) equals the resistance (R).
Since capacitive reactance is defined as XC = 1 / (2πfC), setting XC = R yields:
R = 1 / (2πfcC)
Rearranging for frequency gives us the core formula:
fc = 1 / (2πRC)
| Symbol | Parameter | Base SI Unit | Description |
|---|---|---|---|
| fc | Cutoff Frequency | Hertz (Hz) | The -3dB corner frequency where power halves. |
| R | Resistance | Ohms (Ω) | Series resistance limiting current to the cap. |
| C | Capacitance | Farads (F) | Shunt capacitance storing charge to ground. |
| π | Pi | Dimensionless | Mathematical constant (≈ 3.14159). |
Crucial Assumptions: This formula assumes ideal components (zero parasitic inductance, zero equivalent series resistance in the capacitor) and an unloaded output. If you connect a load resistor across the capacitor, it forms a parallel resistance that alters both the DC gain and the actual cutoff frequency.
Rearranged Forms: Solving for R or C
On the bench, you rarely know both R and C and ask for the frequency. Usually, you have a target cutoff frequency and a bin of standard components. Use these algebraic rearrangements to find the missing part value:
- Solving for Resistance (R):
R = 1 / (2π × fc × C) - Solving for Capacitance (C):
C = 1 / (2π × fc × R)
Worked Examples with Strict Unit Tracking
The most common point of failure in filter math is unit mismanagement. The formula only works if you convert all prefixes (micro, nano, kilo) into base SI units (Farads, Ohms, Hertz) before calculating.
Example 1: Audio Subwoofer Crossover (Target: 80 Hz)
Scenario: You are building a passive line-level crossover and need an 80 Hz low pass filter. You have a high-quality 1 μF film capacitor in your parts bin. What resistor do you need?
- Identify knowns and convert to base units:
fc = 80 Hz
C = 1 μF = 1 × 10-6 F - Select the rearranged formula:
R = 1 / (2π × fc × C) - Substitute and calculate intermediate steps:
Denominator = 2 × 3.14159 × 80 × (1 × 10-6)
Denominator = 502.654 × 10-6 = 0.000502654 - Final Division:
R = 1 / 0.000502654 = 1989.4 Ω
Bench Reality: 1989 Ω is not a standard value. You would use a 2.0 kΩ (E24 series) resistor, which shifts the actual cutoff to roughly 79.5 Hz—well within acceptable audio tolerances.
Example 2: Smoothing a 5 kHz Arduino PWM Signal
Scenario: You are using an Arduino Uno’s 5 kHz PWM pin as a crude DAC. To get a smooth DC voltage, you want the filter’s cutoff frequency to be one decade below the PWM frequency (500 Hz) to achieve -20dB of attenuation on the carrier. You choose a 10 kΩ resistor to avoid loading the GPIO pin. What capacitor do you need?
- Identify knowns and convert:
fc = 500 Hz
R = 10 kΩ = 10,000 Ω - Select the rearranged formula:
C = 1 / (2π × fc × R) - Substitute and calculate:
Denominator = 2 × 3.14159 × 500 × 10,000
Denominator = 31,415,900 - Final Division:
C = 1 / 31,415,900 = 3.183 × 10-8 F - Convert back to engineering prefixes:
3.183 × 10-8 F = 31.83 nF
Bench Reality: The closest standard E12 capacitor is 33 nF. Using 33 nF drops your cutoff slightly to ~482 Hz, which is perfectly fine for smoothing a 5 kHz carrier.
Common Unit Traps That Break Your Math
If your calculated magnitude looks wildly wrong, you likely fell into one of these three traps:
Capacitors are almost never sold in base Farads. A 100 nF capacitor is 0.0000001 F (1 × 10-7 F). If you type "100" into your calculator instead of "100e-9", your calculated resistor value will be off by a factor of 100 million.
Trap 2: Angular vs. Standard Frequency. In advanced AC theory, you will see the formula ωc = 1 / RC. Here, ω (omega) is angular frequency measured in radians per second. If your target is in Hertz (cycles per second), you must use the 2π version of the formula. Forgetting the 2π shifts your cutoff frequency by a factor of 6.28.
Trap 3: Ignoring the Load Impedance. The formula assumes the output is measured by an infinite-impedance device (like a standard oscilloscope at 1MΩ). If you use R = 10 kΩ and drive a load that is also 10 kΩ, the load acts in parallel with the capacitor at DC. Your passband voltage will drop by 50%, and the effective resistance seen by the capacitor changes, shifting fc higher. For low-impedance loads, you must buffer the filter with a unity-gain op-amp like the TI LM358.
Decision Path: Picking Real-World Component Values
Calculating the exact math is only step one. Step two is mapping that math to physical parts that won't cause secondary issues. Use this decision tree when designing a low pass filter for microcontroller PWM smoothing (e.g., ESP32 or Arduino).
| Condition / Constraint | Action | Resulting Pick |
|---|---|---|
| Target Application: Smoothing ESP32 8 kHz PWM to analog DC. | Set target fc to 1/10th of PWM frequency for adequate ripple rejection. | Target fc = 800 Hz |
| Source Constraint: ESP32 GPIO max current is ~40mA, but keep it under 10mA for safety. | Select R ≥ 3.3V / 10mA. Choose a high impedance to protect the pin. | R = 10 kΩ (Standard E24) |
| Calculate C: C = 1 / (2π × 800 × 10,000) | Math yields 1.989 × 10-8 F (19.89 nF). | Ideal C = 19.89 nF |
| Component Availability: 19.89 nF does not exist in standard kits. | Round to the nearest E12 series capacitor value (higher C lowers fc, which is safer for filtering). | Final C = 22 nF (Ceramic X7R) |
| Verify Final fc: 1 / (2π × 10000 × 22e-9) | Ensure the shifted frequency still adequately filters the 8 kHz carrier. | Actual fc = 723 Hz (Perfect) |
Concrete BOM Recommendation: For general microcontroller PWM smoothing where the output feeds a high-impedance ADC or op-amp buffer, default to a 10 kΩ 1/4W metal film resistor and a 22 nF X7R ceramic capacitor. This combination yields a ~723 Hz cutoff, provides excellent protection for the GPIO pin, and uses parts that cost fractions of a cent.
Realistic Magnitudes and When the Ideal Formula Fails
Knowing what a "normal" answer looks like saves you from chasing ghost bugs. Here are the realistic magnitude bands for low pass filters based on application:
- Audio Subwoofers: 20 Hz to 120 Hz. (Requires large resistors or large inductors/capacitors if passive).
- Audio Tweeters (High Pass, inverse concept): 2 kHz to 10 kHz.
- PWM DAC Smoothing: 100 Hz to 1 kHz.
- RF / EMI Snubber Filtering: 1 MHz to 100+ MHz. (At these frequencies, the physical leads of a standard resistor introduce parasitic inductance, breaking the simple RC formula. You must use surface-mount (SMD) 0402 or 0603 components and account for PCB trace capacitance).
When to abandon the passive RC formula: If you need a steep rolloff (e.g., cutting off everything above 1 kHz with a -40dB/decade slope instead of the -20dB/decade slope of a single RC stage), a passive filter will require cascading stages that load each other down, ruining your math. In this case, transition to an active Sallen-Key topology using an op-amp. The Analog Devices Filter Wizard is the industry standard tool for calculating component values for these higher-order active filters.
For 90% of bench tasks—cleaning up a noisy sensor reading, dropping PWM to analog, or rolling off high-frequency hiss—the first-order RC formula fc = 1 / (2πRC) is all you need. Convert your units to base SI, pick the nearest E-series standard values, and verify with an oscilloscope.
