A low-pass filter is a circuit that allows low-frequency signals to pass through to the output while attenuating the amplitude of frequencies above a specific threshold known as the cutoff frequency. In a real circuit, what this component network actually changes is the frequency spectrum of your signal—it strips out high-frequency electromagnetic interference (EMI), smooths pulsed digital outputs into stable analog DC voltages, or directs bass frequencies to a subwoofer while blocking treble. It does not act as a brick wall; instead, it gradually rolls off unwanted high frequencies at a predictable rate.
Standard Component Pairings and Roll-Off Rates
The most common implementation is the first-order passive RC (resistor-capacitor) low-pass filter. In this topology, the resistor is placed in series with the signal path, and the capacitor is placed in parallel from the output node to ground. Think of the capacitor like a flexible water bladder attached to a pipe, where the resistor is a narrow section of that pipe. If you rapidly pulse the water pressure (high frequency), the narrow pipe restricts flow and the bladder doesn't have time to expand—the pressure wave is blocked. If you slowly change the pressure (low frequency), the bladder has time to fill and push the water through to the other side.
Below is a reference table of standard 1st-order RC configurations. These values use standard E12/E24 component series and are calculated using the standard formula $f_c = 1 / (2\pi RC)$.
| Resistor (R) | Capacitor (C) | Cutoff Freq ($f_c$) | Roll-Off Rate | Primary Application |
|---|---|---|---|---|
| 1 kΩ | 100 nF | 1,591 Hz | -20 dB/decade | General audio tone control / treble bleed |
| 10 kΩ | 100 nF | 159 Hz | -20 dB/decade | PWM to analog DAC smoothing (e.g., Arduino/ESP32) |
| 10 kΩ | 10 nF | 1,591 Hz | -20 dB/decade | High-frequency EMI snubber / I2S data line filtering |
| 100 Ω | 1 μF | 1,591 Hz | -20 dB/decade | Low-impedance power supply ripple filtering |
| 4.7 kΩ | 47 nF | 720 Hz | -20 dB/decade | Sensor noise filtering (thermistors / potentiometers) |
The Math: Calculating Cutoff and Output Voltage
Let's walk through a concrete numeric example using a very common microcontroller scenario: smoothing a 5V PWM signal from an ESP32 GPIO pin into a readable analog DC voltage. We will use a 10 kΩ resistor and a 100 nF capacitor.
Step 1: Find the Cutoff Frequency
The formula for the -3dB cutoff frequency is:
$f_c = \frac{1}{2 \pi R C}$
Plugging in our real values (R = 10,000 Ω, C = 0.0000001 F):
$f_c = \frac{1}{2 \times 3.14159 \times 10000 \times 0.0000001}$
$f_c = \frac{1}{0.00628318} \approx 159.15 \text{ Hz}$
Step 2: Calculate Output Voltage at Specific Frequencies
If we feed a steady 5.0V DC signal (0 Hz) into this filter, the capacitor charges fully, and the output is 5.0V. But what happens at the cutoff frequency of 159.15 Hz?
At $f_c$, the output voltage drops to 70.7% of the input:
$V_{out} = 5.0V \times 0.707 = \mathbf{3.535V}$
What happens one decade higher, at 1,591 Hz? A first-order filter rolls off at -20 dB per decade. A -20 dB drop means the voltage is reduced to exactly 1/10th of the input:
$V_{out} = 5.0V \times 0.1 = \mathbf{0.5V}$
When building analog filters for precision sensor conditioning, never use standard X7R or Y5V ceramic capacitors. These dielectrics exhibit severe DC bias characteristics—a 100nF X7R cap might drop to 40nF when 5V is applied across it, shifting your cutoff frequency up by 150%. Always specify C0G/NP0 dielectrics for ceramic capacitors under 1μF, or use film capacitors (like the WIMA MKS series) for larger values to maintain an accurate $f_c$.
Where You Meet Low-Pass Filters in Practice
You will rarely build a filter just for the sake of theory. Here is where passive low-pass filters earn their keep on the bench and in the field.
1. PWM to Analog Conversion (DAC Smoothing)
Microcontrollers like the Arduino Uno lack a true Digital-to-Analog Converter (DAC). Instead, they output Pulse Width Modulation (PWM). If you measure a 5kHz PWM signal with a multimeter, it reads an average voltage, but an oscilloscope reveals a harsh square wave switching between 0V and 5V. By placing an RC low-pass filter (e.g., 10kΩ and 1μF, $f_c \approx 15.9$ Hz) at the output, the capacitor averages the pulses, yielding a smooth, flat DC voltage proportional to the duty cycle. As of 2026, while chips like the ESP32-S3 feature internal DACs, RC filtering remains the cheapest way to add extra analog outputs to any digital pin.
2. Sensor Signal Conditioning
Analog sensors like thermistors, photoresistors, and potentiometers act as high-impedance voltage dividers. They are highly susceptible to picking up high-frequency RF noise from nearby switching power supplies or Wi-Fi antennas. A simple low-pass filter with a cutoff around 10 Hz to 50 Hz placed right at the microcontroller's ADC pin acts as an anti-aliasing filter, stripping out the noise and preventing the ADC from sampling garbage data. For deeper reading on how microcontrollers interpret these smoothed signals, review SparkFun's Guide to Pulse Width Modulation and ADC sampling techniques.
3. Audio Crossovers and Tone Control
In speaker building, inductors and capacitors form passive crossover networks. A low-pass filter (using an inductor in series and a capacitor in parallel) ensures that deep bass frequencies reach the subwoofer, while high frequencies are shunted to ground before they can damage the woofer's voice coil or cause harsh distortion.
Common Confusions and Troubleshooting Mistakes
Even experienced hobbyists make predictable mistakes when implementing passive filters. Here is what people commonly confuse or misunderstand about low-pass filter behavior.
Confusion 1: "The cutoff frequency blocks everything above it."
Reality: A passive filter is a slope, not a cliff. At the cutoff frequency, 70.7% of the voltage still passes through. To get a steeper "brick wall" effect, you must cascade multiple stages to create a 2nd-order (-40 dB/decade) or 3rd-order (-60 dB/decade) filter, or switch to an active filter using an op-amp.
Confusion 2: "I can use any resistor and capacitor combination that equals my target $f_c$."
Reality: The math $R \times C$ might be identical for a 1MΩ/10pF pair and a 10Ω/1μF pair (both yield ~15.9 kHz), but the impedance is entirely different. A 1MΩ resistor makes the filter extremely susceptible to noise pickup and loading effects. Always choose the lowest practical resistor value that your driving source can handle without exceeding its current limits, and scale the capacitor up accordingly.
Confusion 3: "My filtered DC voltage is lower than my math predicted."
Reality: You are experiencing loading effects. The output impedance of a passive RC filter is roughly equal to the resistor value at low frequencies. If your 10kΩ filter feeds into an ADC or a multimeter with a 10kΩ input impedance, you have accidentally created a 50% voltage divider. The fix is to either lower the filter's resistor value (and increase the capacitor) or place a unity-gain op-amp buffer between the filter and the load.
For a deeper theoretical breakdown of how passive components interact with AC impedance, the Electronics Tutorials guide on Passive Low-Pass Filters provides excellent Bode plot visualizations. Understanding the relationship between resistance, capacitance, and frequency is foundational for anyone moving from simple digital blinking to mixed-signal circuit design.
