A lowpass RC filter is a passive circuit consisting of a resistor and a capacitor that allows low-frequency signals to pass through while attenuating high-frequency signals. In a real circuit, it changes both the amplitude and the phase of the AC components of a signal, introducing a frequency-dependent voltage drop and a phase lag that reaches exactly -45° at the cutoff frequency. Think of it like a car's suspension system: the spring (capacitor) and damper (resistor) allow the vehicle to smoothly follow slow, rolling hills (low frequencies) while absorbing and isolating the cabin from sharp, high-frequency gravel vibrations.
The Core Math and Standard Component Cutoff Table
The behavior of a passive resistor-capacitor network is defined by its cutoff frequency ($f_c$), also known as the corner frequency. This is the exact point where the output power drops by half, or equivalently, the output voltage drops to 70.7% of the input voltage.
When designing on the bench, you rarely calculate from scratch. Instead, you select standard E12 or E24 resistor and capacitor values to hit a target frequency band. The table below provides exact cutoff frequencies for common, easily sourced component pairings. Keep this in your lab notebook for rapid prototyping.
| Resistor (R) | Capacitor (C) | Cutoff Freq ($f_c$) | Primary Use Case |
|---|---|---|---|
| 1 kΩ | 100 nF | 1,592 Hz | Audio line-level filtering, basic noise removal |
| 10 kΩ | 100 nF | 159 Hz | Subwoofer crossovers, slow sensor averaging |
| 10 kΩ | 10 nF | 1,592 Hz | Higher impedance audio stages, anti-aliasing |
| 100 kΩ | 10 nF | 159 Hz | Low-power battery-operated sensor filtering |
| 100 kΩ | 100 nF | 15.9 Hz | Thermocouple / thermistor DC smoothing |
For a deeper theoretical breakdown of the complex impedance math behind these values, the All About Circuits textbook chapter on passive filters provides excellent AC phasor diagrams.
Worked Example: Smoothing a 5V PWM Signal to Analog DC
Microcontrollers like the ESP32 or Arduino Uno do not have true analog voltage outputs; they simulate them using Pulse Width Modulation (PWM). If you feed a 5V PWM signal at a 50% duty cycle into a multimeter, it will read an average of 2.5V. However, an oscilloscope will show a square wave violently swinging between 0V and 5V. To convert this into a steady DC voltage for driving an analog actuator or an audio amplifier, we use a lowpass RC filter.
The Scenario: We have an ESP32 outputting a 20 kHz PWM signal (a common default for LEDC peripherals) at 50% duty cycle. We want a clean 2.5V DC output with minimal ripple.
Step 1: Determine the Target Cutoff Frequency
A standard rule of thumb for PWM smoothing is to set the filter's cutoff frequency to at least one-tenth of the PWM frequency to adequately suppress the fundamental switching harmonic.
$f_{target} = \frac{20,000 \text{ Hz}}{10} = 2,000 \text{ Hz}$
Step 2: Select the Capacitor
Let's choose a 1 µF capacitor. Bench Tip: For PWM smoothing where precision isn't critical, a standard X7R ceramic capacitor is fine. However, if you are filtering an audio signal, you must use a C0G/NP0 dielectric capacitor. X7R ceramics exhibit piezoelectric microphonics and voltage coefficients that will inject distortion into audio paths.
Step 3: Calculate the Resistor
Rearranging the cutoff formula to solve for R:
$R = \frac{1}{2 \pi f_c C}$
$R = \frac{1}{2 \pi \times 2000 \times 1 \times 10^{-6}} = 79.57 \text{ } \Omega$
The nearest standard E12 resistor value is 82 Ω.
Step 4: Verify Actual Cutoff and Ripple
Using the 82 Ω resistor and 1 µF capacitor, our actual cutoff frequency is:
$f_c = \frac{1}{2 \pi \times 82 \times 10^{-6}} = 1,941 \text{ Hz}$
To estimate the peak-to-peak ripple voltage ($V_{ripple}$) remaining on our DC line, we use the approximation for a 50% duty cycle square wave:
$V_{ripple(p-p)} \approx \frac{V_{in}}{2 \times f_{pwm} \times R \times C}$
$V_{ripple(p-p)} \approx \frac{5}{2 \times 20000 \times 82 \times 10^{-6}} = \frac{5}{3.28} \approx 1.52 \text{ V}$
A 1.52V ripple is still quite large for sensitive analog circuits. To fix this without dropping the DC voltage level, you would either increase the PWM frequency to 100 kHz, increase the capacitor to 10 µF (which drops ripple to ~0.15V), or cascade a second identical RC stage. For a comprehensive look at how ADCs interpret these signals, review the Wikipedia entry on anti-aliasing and reconstruction filters.
Where You Meet the Lowpass RC Filter in Practice
You will encounter this exact topology across almost every domain of electronics. Recognizing the application dictates your component tolerances.
- Sensor Noise Filtering (Thermistors & LDRs): Analog sensors read environmental data that is inherently slow-changing. A 100 kΩ resistor and a 100 nF capacitor ($f_c = 15.9$ Hz) placed right at the ADC pin will short out high-frequency EMI picked up by the sensor wires without delaying the actual temperature reading.
- Anti-Aliasing Before ADCs: According to the Nyquist-Shannon sampling theorem, any frequency above half your ADC's sample rate will fold back into your data as false low-frequency noise. A lowpass RC filter is placed before the ADC to physically chop off these high frequencies. For high-precision ADCs (like the ADS1115), this filter is often followed by an op-amp buffer to drive the ADC's internal sampling capacitor.
- Audio Crossovers: In passive speaker crossovers, a lowpass filter routes low frequencies to the woofer while blocking high frequencies that could overheat the woofer's voice coil. (Note: High-power audio crossovers typically use LC filters rather than RC, to avoid burning watts as heat in the resistor).
Common Confusions and Mistakes to Avoid
The most common mistake hobbyists make is forgetting that a passive RC filter's output impedance is determined by the resistor. If you build a filter with a 10 kΩ resistor and connect it to a microcontroller ADC pin that has a 10 kΩ input impedance, you have accidentally created a voltage divider. Your DC voltage will drop by 50%, and your cutoff frequency will shift. Always ensure the load impedance is at least 10x to 100x greater than the filter resistor, or use a unity-gain op-amp buffer.
The "Brick Wall" Myth
Beginners often assume that "cutoff frequency" means all frequencies above that point are completely blocked. This is false. A single-pole RC filter rolls off at a gentle -20 dB per decade (or -6 dB per octave). If your cutoff is 1 kHz, a 10 kHz signal is not eliminated; it is merely attenuated by about -20 dB (reduced to 10% of its original voltage). If you need a steeper "brick wall" drop-off, you must cascade multiple RC stages or switch to an active Sallen-Key op-amp topology.
RC vs. LC Filters
People frequently confuse when to use an RC filter versus an LC (inductor-capacitor) filter. Use RC for low-current signal processing where power loss in the resistor is negligible. Use LC for power supply filtering or high-current audio crossovers. Inductors block high frequencies without dissipating real power as heat, but they are physically larger, more expensive, and can introduce unwanted magnetic coupling or resonance ringing (Q-factor peaking) if not properly damped.
Frequently Asked Questions
Can I put the capacitor first and the resistor second?
No. In a standard lowpass configuration, the resistor must be in series with the signal path, and the capacitor must be in parallel (shunting to ground). If you swap them, you create a highpass filter, which blocks DC and passes high frequencies.
Why does my RC filter output look like a triangle wave instead of DC?
This happens when the RC time constant ($\tau = R \times C$) is too small relative to the input frequency. The capacitor is charging and discharging linearly during each PWM cycle because it never reaches saturation. Increase the capacitor value to increase the time constant.
Do I need to worry about capacitor polarity?
If you are using ceramic, film, or mica capacitors, they are non-polarized and can be placed in either direction. If your design requires large values (e.g., >10 µF) and you must use an electrolytic or tantalum capacitor, ensure the positive anode is tied to the DC bias voltage, and the signal never swings below ground, or the capacitor will fail catastrophically.
