A 1st order filter is an electronic circuit that uses a single reactive component—either one capacitor or one inductor—paired with a resistor to attenuate unwanted frequencies at a steady rate of 20 dB per decade. In a real circuit or installation, it changes the signal's frequency profile by smoothly rolling off high-frequency noise (in a low-pass configuration) or blocking DC drift (in a high-pass configuration), all without introducing the complex phase shifts, component costs, or resonance ringing associated with higher-order active filters. Builders commonly confuse the -3dB cutoff frequency with a "brick-wall" stopband, assuming all frequencies above this point are completely blocked; in reality, the filter merely begins its gradual attenuation at this threshold.
The Math and Mechanics: How the Roll-Off Works
At the heart of every passive 1st order filter is a voltage divider where one of the resistive elements is replaced by a frequency-dependent reactive component. In the most common variant—the RC (resistor-capacitor) low-pass filter—the resistor sits in series with the signal path, and the capacitor sits in parallel to ground. As detailed in SparkFun's RC Filter Tutorial, the capacitor's impedance drops as frequency rises, effectively shorting high-frequency AC signals to ground while letting DC and low-frequency signals pass through to the output.
The defining characteristic of this topology is its -20 dB/decade roll-off. This means that for every tenfold increase in frequency beyond the cutoff point ($f_c$), the signal's voltage amplitude drops to one-tenth of its previous value. The cutoff frequency, also known as the corner frequency, is calculated using the formula:
$f_c = \frac{1}{2 \pi R C}$
At exactly $f_c$, the output voltage drops to 70.7% of the input voltage. In power terms, this is a 50% reduction, which translates to the famous -3dB point. Because the slope is relatively gentle, a 1st order filter is rarely used when you need aggressive noise rejection in a tight frequency band, but it excels in applications where phase linearity and stability are paramount.
Worked Numeric Example: Smoothing an ESP32 PWM Signal
Let’s apply this to a common bench scenario: you are using an ESP32 to generate a pseudo-analog DC voltage via Pulse Width Modulation (PWM) to drive the analog control pin of a motor controller. The ESP32's default PWM frequency is 1 kHz, resulting in a square wave that swings between 0V and 3.3V. The motor controller needs a smooth DC voltage, not a 1 kHz square wave.
We need a low-pass 1st order filter to extract the DC average. A good rule of thumb is to set the cutoff frequency one decade below the fundamental frequency you want to eliminate. Therefore, our target $f_c$ is 100 Hz.
- Choose the Resistor: We select a 10 kΩ resistor. This is high enough to avoid drawing excessive current from the ESP32 GPIO pin (which has a max source current of ~40mA, but practically should be kept under 12mA), but low enough to overcome the input impedance of the motor controller.
- Calculate the Capacitor: Rearranging the formula to solve for C:
$C = \frac{1}{2 \pi \times R \times f_c}$
$C = \frac{1}{2 \pi \times 10,000 \times 100} = 0.000000159 \text{ Farads}$, or 159 nF. - Select a Standard Value: 159 nF is not a standard E12 component value. The nearest standard values are 150 nF and 220 nF. We will choose 220 nF to ensure a slightly lower cutoff frequency for better ripple rejection.
- Verify the Real-World Cutoff: Plugging 220 nF back into the formula gives us an actual $f_c$ of 72.3 Hz.
What does this mean for the 1 kHz ripple? At 1 kHz, the frequency is roughly 13.8 times higher than our 72.3 Hz cutoff. A 1st order filter attenuates the voltage by a factor of $\frac{1}{\sqrt{1 + (f/f_c)^2}}$. This results in an attenuation factor of 0.072 (or -22.8 dB). If your 3.3V PWM signal has a 3.3V peak-to-peak ripple, the filter reduces that ripple to roughly 237 mV peak-to-peak riding on top of your DC average. For a motor controller, this is perfectly acceptable. For a precision 16-bit ADC, it would not be, highlighting exactly when you must upgrade to a 2nd order active filter.
For this motor control application, a cheap X7R ceramic 220nF capacitor is fine. However, if you were building an anti-aliasing filter for a precision load cell or RTD sensor, you must use a C0G/NP0 dielectric capacitor. X7R capacitors exhibit voltage coefficient (capacitance drops as applied DC voltage rises) and microphonic noise, which will introduce non-linear distortion into precision analog measurements.
Where You Meet 1st Order Filters in Practice
While they lack the aggressive stopband rejection of higher-order topologies, 1st order filters are ubiquitous in both electronics and electrical installations due to their inherent stability and low cost.
- Sensor Signal Conditioning: Thermocouples and RTDs produce slow-moving DC signals that pick up high-frequency RFI from nearby VFDs or radio transmitters. A simple 1st order RC low-pass filter at the instrumentation amplifier input prevents this RF energy from being rectified into a false DC offset.
- Audio Crossovers: The series capacitor placed in front of a piezo tweeter in a basic speaker cabinet is a 1st order high-pass filter. It blocks the low-frequency bass energy that would otherwise overheat and destroy the tweeter's voice coil.
- Power Supply Ripple Reduction: A 1st order RL (resistor-inductor) or RC filter is often placed after a bridge rectifier to smooth out the 120 Hz (or 100 Hz) mains ripple before it reaches a linear voltage regulator.
- EMI Snubbers: In electrical panels, a series RC network placed across a contactor coil acts as a specialized high-frequency low-pass filter, absorbing the massive inductive voltage spike (back-EMF) when the coil is de-energized, protecting sensitive PLC inputs on the same bus.
1st Order vs. Higher-Order Filters: When to Stop at One
Deciding between a 1st order filter and a 2nd (or higher) order filter is a trade-off between simplicity and stopband aggression. As outlined in resources like Electronics Tutorials on Filter Design, adding more poles increases the roll-off rate but introduces new failure modes.
| Criteria | 1st Order Filter | 2nd Order Filter (e.g., Sallen-Key) | 3rd Order+ (e.g., Butterworth/Chebyshev) |
|---|---|---|---|
| Roll-Off Rate | -20 dB/decade | -40 dB/decade | -60 dB/decade or steeper |
| Phase Shift at $f_c$ | -45° | -90° | -135° or more |
| Component Count | 1 R, 1 C (or L) | 2 R, 2 C, plus Op-Amp | Multiple R, C, and Op-Amps |
| Ringing / Resonance Risk | None (overdamped) | Moderate (depends on Q factor) | High (requires precise tuning) |
| Best Used When... | Phase linearity is critical, or noise is far from the signal band. | Noise is close to the signal band and needs aggressive cutting. | Strict anti-aliasing or RF band-limiting is required. |
A 1st order filter never rings or overshoots in the time domain. If you apply a step input (like a sudden voltage spike), a 1st order filter will smoothly and exponentially approach the new voltage. A poorly tuned 2nd order filter will overshoot the target voltage and oscillate (ring) before settling, which can trigger false logic thresholds in downstream digital circuits.
Frequently Asked Questions
Why is the cutoff frequency of a 1st order filter called the -3dB point?
The decibel (dB) scale is logarithmic. In terms of power, a -3dB drop means the power has been exactly halved (50%). Because power is proportional to voltage squared ($P = V^2/R$), halving the power requires the voltage to drop to $\frac{1}{\sqrt{2}}$, which is approximately 0.707 (or 70.7%) of the input voltage. Therefore, the frequency at which the output voltage reaches 70.7% of the input is designated as the -3dB cutoff point.
Can I cascade two passive 1st order filters to make a 2nd order filter?
Technically yes, but practically it rarely works as intended due to loading effects. If you connect two identical passive RC low-pass filters directly in series, the second stage loads the first stage, altering the impedance network. The resulting circuit will not yield a clean -40dB/decade Butterworth response; instead, it will have a heavily damped, sagging response curve. To properly cascade passive filters, you must insert a unity-gain buffer op-amp between the stages to isolate their impedances, effectively turning it into an active filter.
What do people commonly confuse a 1st order filter with?
The most common confusion is mistaking the -3dB cutoff frequency for a "brick-wall" boundary. Many beginners assume that a 100 Hz low-pass filter completely eliminates a 120 Hz signal. In reality, because the roll-off is only -20dB/decade, a 120 Hz signal will pass through a 100 Hz 1st order filter with very little attenuation (only about -1.6 dB). If you need to heavily attenuate a frequency that is close to your passband, a 1st order filter is mathematically insufficient, and you must use a higher-order topology.
Should I use an RC or RL 1st order filter for my circuit?
RC (resistor-capacitor) filters are the default choice for 95% of low-voltage signal applications because capacitors are cheap, small, and do not emit magnetic fields. RL (resistor-inductor) filters are generally reserved for high-current power applications, such as filtering the output of a switching buck converter. Inductors can handle high DC currents without the massive power dissipation (heat) that a series resistor would cause in an RC configuration. However, inductors are bulky, expensive, and can radiate EMI if not properly shielded.
