An active low pass filter circuit op amp design using the Sallen-Key topology provides a sharp -40dB/decade roll-off without the insertion loss and low input impedance of passive RC networks. For a standard 1kHz audio cutoff with a Butterworth response, use two 11kΩ resistors, a 20nF capacitor, and a 10nF capacitor paired with an OPA2134 or TL072 dual op-amp. This configuration yields a high-impedance input, a low-impedance output, and unity gain in the passband.
Why Active Over Passive? Topology & Node Labels
A simple passive RC filter suffers from a shallow -20dB/decade roll-off and loads down whatever circuit follows it. By wrapping a second RC stage in an op-amp feedback loop, we create a second-order active filter with a -40dB/decade slope. While the Multiple Feedback (MFB) topology is an alternative, the Sallen-Key (unity-gain) topology wins for 90% of hobbyist and sensor-aliasing applications because it is non-inverting, inherently stable at unity gain, and requires no precise resistor matching to set the DC gain.
To understand the signal flow, map out these specific node labels on your schematic:
- Vin: The raw signal input.
- Node A: The junction between R1 and R2. This is also where the feedback capacitor (C1) connects to the output.
- Node B: The junction between R2 and the non-inverting input (Pin 3 on a standard DIP-8). This is also where the grounding capacitor (C2) connects.
- Vout: The op-amp output (Pin 6), which feeds back to C1 and drives the next stage.
Design Walkthrough: Picking Real Component Values
Let's design a 1kHz Butterworth low pass filter. A Butterworth response gives us a maximally flat passband (Quality factor Q = 0.707) with no peaking at the cutoff frequency. For a unity-gain Sallen-Key Butterworth filter, the component ratios are strictly defined: R1 = R2 = R, and C1 = 2 × C2.
The cutoff frequency formula is:
f_c = 1 / (2π × R × √(C1 × C2))
Step 1: Pick the capacitors. Capacitor values dictate the impedance scale of the filter. We want impedances in the 1kΩ–100kΩ range to avoid excessive thermal noise (too high) or op-amp drive limitations (too low). Let's select C2 = 10nF. Following the Butterworth ratio, C1 = 20nF. (On the bench, build C1 using two 10nF C0G/NP0 capacitors in parallel).
Step 2: Calculate the resistors. Plugging our values into the rearranged formula:
R = 1 / (2π × 1000 × √(20e-9 × 10e-9))
R ≈ 11,253 Ω
Step 3: Select standard E24 values. The closest standard 1% resistor value is 11.0kΩ. Recalculating with 11kΩ gives an actual cutoff of ~1,018 Hz, which is well within standard component tolerances.
Step 4: Choose the Op-Amp and Dielectric. Use an OPA2134 or TL072 for audio applications. Crucially, you must use C0G/NP0 dielectric ceramic capacitors or film capacitors. Never use X7R or Y5V ceramics for filter networks; their capacitance drops significantly under applied DC bias voltage and they exhibit microphonic piezoelectric effects, which will introduce severe harmonic distortion into your audio or sensor signals.
Component Behavior & Failure Mode Analysis
Understanding how each component shapes the transfer function is critical for debugging. Below is a behavior matrix detailing what happens when you alter a single element, followed by the extreme failure modes.
| Component | Change | Effect on Cutoff (f_c) | Effect on Q (Peaking) |
|---|---|---|---|
| R1 or R2 | Increase Resistance | Decreases (shifts left) | Minor shift, maintains Butterworth if matched |
| C1 (Feedback) | Increase Capacitance | Decreases (shifts left) | Decreases Q (overdamped, sluggish roll-off) |
| C2 (Ground) | Increase Capacitance | Decreases (shifts left) | Increases Q (causes passband peaking/ringing) |
What Breaks at the Extremes?
When troubleshooting a dead or misbehaving breadboard, check these specific failure modes:
- Short C1: The feedback path is destroyed. The circuit degrades into a basic 1st-order passive RC filter followed by a buffer. You lose the -40dB/decade slope, dropping to -20dB/decade.
- Open C2: Node B loses its AC ground path. While DC might pass, the op-amp's input bias current will slowly charge the parasitic capacitance at Node B, eventually railing the op-amp output to the positive or negative supply voltage.
- Short R2: Node A and Node B merge. C1 and C2 are now effectively in parallel to ground from the perspective of R1. The circuit becomes a 1st-order filter.
- Open R1: No signal reaches the op-amp input. The output will sit at 0V (in a dual-supply system) or rail due to unbalanced bias currents.
Breadboard Testing: Step-by-Step Verification
Breadboarding high-gain or high-frequency analog circuits invites parasitic oscillation. Follow this exact sequence to verify your low pass filter circuit op amp design without chasing ghost signals.
- Power and Decouple: Apply ±12V (or ±9V) to the op-amp rails. Place a 100nF MLCC decoupling capacitor directly across the V+ and V- pins (e.g., Pins 8 and 4 on a DIP-8) with the shortest possible lead dress. Do not skip this, or the op-amp will act as an RF oscillator.
- Wire the DC Path First: Install the op-amp, R1, R2, and the jumper from Vout to the inverting input (Pin 2) to set unity gain. Leave C1 and C2 out for a moment.
- Verify the Buffer: Apply a 1kHz, 1Vpp sine wave to Vin. Probe Vout with your oscilloscope. You should see exactly 1Vpp. If it's oscillating or railing, check your power pins and ground connections.
- Add the Capacitors: Power down, insert C1 (Node A to Vout) and C2 (Node B to GND). Ensure C0G/NP0 caps are used.
- Sweep and Measure: Power up. Set your function generator to output 1Vpp. Sweep the frequency from 10Hz up to 10kHz. At 1kHz, your scope should read ~0.707Vpp (the -3dB point). At 10kHz (one decade up), the signal should be attenuated by roughly 40dB (down to ~10mVpp).
Frequently Asked Questions
Can I use an LM358 for an audio low pass filter circuit op amp?
While the LM358 is cheap and ubiquitous, it is a poor choice for audio low pass filters. The LM358 uses a Class-B output stage that suffers from severe crossover distortion when the signal crosses 0V. This injects high-frequency harmonics into your audio path, which defeats the purpose of a clean filter. For audio, spend the extra $0.50 on a TL072, NE5532, or OPA2134, which feature Class-AB push-pull outputs and vastly lower noise floors. The LM358 is perfectly fine for slow-moving DC sensor signals (like thermocouples or load cells) where audio-band crossover distortion is irrelevant.
Why is my active low pass filter oscillating on the breadboard?
Oscillation in a unity-gain Sallen-Key filter almost always stems from three issues: inadequate power supply decoupling, excessively long ground leads creating inductive feedback, or driving a highly capacitive load (like a long coaxial cable) directly from the op-amp output. To fix this, ensure 100nF decoupling caps are tight to the IC pins, keep the ground path for C2 as short as physically possible, and add a 47Ω to 100Ω series resistor directly at the op-amp's Vout pin before it hits the breadboard rail or external cable. This isolates the op-amp's internal compensation from the external capacitive load.
How do I change a Sallen-Key low pass filter to a high pass?
The Sallen-Key topology is beautifully symmetrical. To convert your low pass filter into a high pass filter with the exact same cutoff frequency and Q factor, simply swap the positions of the resistors and capacitors. R1 and R2 become C1 and C2, while C1 and C2 become R1 and R2. The node structure remains identical, but the signal path now blocks DC and passes high frequencies through the capacitive voltage divider into the op-amp's non-inverting input. For deeper mathematical proofs on active filter transformations, refer to standard analog design references like the All About Circuits active filter guides or Electronics Tutorials.
