A second order Sallen Key low pass filter is an active RC circuit topology that uses an operational amplifier configured as a unity-gain buffer or non-inverting amplifier to attenuate frequencies above a specific cutoff point with a -40 dB/decade roll-off. In a real circuit or installation, this topology fundamentally changes your signal chain by providing a sharp transition band between your passband and stopband without loading down the source signal. This prevents high-frequency noise from aliasing into ADCs or causing intermodulation distortion in audio chains, all while maintaining a high input impedance and low output impedance.

Think of the two cascaded RC stages as a two-gate security checkpoint where the first guard slows down the crowd, and the second guard turns away anyone still moving too fast, while the op-amp acts as a powered walkway that prevents the second guard from interfering with the first guard's line of sight. Below, we break down the exact math, component selections, and practical gotchas you need to build one on the bench.

Filter Alignments and Component Ratios

Before you pick up a soldering iron, you must choose your filter alignment. The 'shape' of the filter's frequency and phase response is dictated by the damping factor (or Quality factor, Q). Most hobbyists default to Butterworth, but it is rarely the correct choice for time-domain applications like sensor pulse detection. The table below maps the standard alignments to their required component ratios for a unity-gain Sallen Key topology (where $R_1 = R_2 = R$).

Alignment Quality Factor (Q) Damping Factor ($\zeta$) Capacitor Ratio (C1/C2) Best Application
Butterworth 0.707 0.707 2.0 : 1 Audio crossovers, flat magnitude response
Bessel 0.577 0.866 2.6 : 1 Pulse/step signals, DAC reconstruction, minimal phase shift
Chebyshev (0.5dB ripple) 0.864 0.579 1.5 : 1 Steep anti-aliasing where passband ripple is acceptable
Critically Damped 0.500 1.000 3.0 : 1 Control loops, avoiding any overshoot in step response
Bench Tip: If you are designing for audio, always use the Butterworth alignment for its maximally flat passband. If you are filtering a load cell or a PWM DAC output where you care about the shape of a square wave or step response, use Bessel. A Butterworth filter will cause visible 'ringing' on the edges of a square wave.

Worked Numeric Example: 1 kHz Butterworth Design

Let's design a unity-gain, second order Sallen Key low pass filter with a cutoff frequency ($f_c$) of 1 kHz using a Butterworth alignment. We will constrain our design to standard off-the-shelf E24 resistor and E12 capacitor values, which is the reality of most workbench builds.

Step 1: Define the governing equations.
For a unity-gain Sallen Key filter where $R_1 = R_2 = R$, the cutoff frequency is defined as:

$$f_c = \frac{1}{2\pi R \sqrt{C_1 C_2}}$$

For a Butterworth response, we established in the table above that $C_1 = 2 \times C_2$.

Step 2: Select the capacitors.
Capacitors are harder to source in exact values than resistors, so we always pick them first. Let's choose a standard E12 value for $C_2$ of 10 nF.
Therefore, $C_1$ must be $2 \times 10\text{ nF} = 20\text{ nF}$. The closest standard E12 value is 22 nF. We will use 22 nF and accept a slight shift in our Q factor, or we can parallel two 10 nF caps for exactness. Let's assume we parallel two 10 nF C0G caps for an exact 20 nF.

Step 3: Calculate the resistors.
Substitute $C_1 = 20\text{ nF}$ and $C_2 = 10\text{ nF}$ into the frequency equation and solve for R:

$$1000 = \frac{1}{2\pi \times R \times \sqrt{20\times 10^{-9} \times 10\times 10^{-9}}}$$

$$1000 = \frac{1}{2\pi \times R \times 1.414 \times 10^{-8}}$$

$$R = \frac{1}{8.885 \times 10^{-5}} \approx 11,254\ \Omega$$

Step 4: Map to E24 values and verify.
The closest standard E24 resistor value is 11 kΩ (or 11.3 kΩ if using E96 1% resistors). Using 11 kΩ, our actual cutoff frequency shifts slightly to 1,023 Hz, which is well within the tolerance of standard audio and sensor applications.

Component Warning: Never use X7R or Y5V ceramic capacitors for the $C_1$ and $C_2$ positions in a Sallen Key filter. These dielectrics exhibit severe voltage coefficients (capacitance drops as voltage increases) and piezoelectric microphonics. Always specify C0G/NP0 ceramics or polypropylene film capacitors for filter stages. See the Analog Devices MT-222 Tutorial for a deep dive on dielectric absorption in active filters.

Where You Meet This in Practice (And What It Isn't)

You will encounter the second order Sallen Key low pass filter most frequently in three specific areas of electronics design:

  • ADC Anti-Aliasing: Placed directly in front of a Successive Approximation Register (SAR) ADC to chop off high-frequency RF noise that would otherwise fold back into the audio or measurement band during sampling.
  • Active Audio Crossovers: Used in powered studio monitors to route low frequencies to the woofer. The unity-gain Sallen Key is favored here because it doesn't introduce phase inversion.
  • Sensor Signal Conditioning: Filtering the high-frequency switching noise out of a strain gauge or thermocouple amplifier before the signal hits a microcontroller.

The Common Confusion: Sallen Key vs. Multiple Feedback (MFB)

The most frequent mistake hobbyists and junior engineers make is confusing the Sallen Key topology with the Multiple Feedback (MFB) low pass filter. While both use one op-amp and achieve a second-order roll-off, their real-world behaviors are vastly different.

In an MFB filter, the op-amp is configured in an inverting topology. This makes the MFB filter highly sensitive to the op-amp's Gain-Bandwidth Product (GBP). If you try to build a 100 kHz MFB filter using a standard TL072 (3 MHz GBP), the filter will fail to achieve the desired Q and will likely oscillate or peak wildly. The Sallen Key, particularly in the unity-gain configuration, is vastly more forgiving of op-amp bandwidth limitations because the op-amp is just acting as a buffer outside the primary RC feedback loop. For high-Q or high-frequency designs, MFB is mathematically superior, but for 90% of bench and audio applications under 50 kHz, the Sallen Key is the more robust, forgiving choice. For more on op-amp bandwidth limitations in active filters, refer to the Texas Instruments SLOA049 Application Note.

Frequently Asked Questions

Do I need a dual-supply op-amp for a Sallen Key filter?
No, but if you are using a single-supply op-amp (like an LM358 or MCP6002), you must bias the non-inverting input to a mid-rail voltage (e.g., VCC/2) using a resistor divider. If you don't, the negative half of your AC signal will be hard-clipped at 0V, destroying the filter's response and introducing massive harmonic distortion.

Why is my Sallen Key filter ringing or oscillating on the breadboard?
Breadboard parasitics. The stray capacitance between breadboard rows (typically 2pF to 5pF) adds directly to your filter capacitors. If you are designing a high-frequency filter (e.g., >100 kHz) using small capacitor values like 100pF, the breadboard's parasitic capacitance will completely detune your Q factor. Move the circuit to a soldered perfboard or PCB with a solid ground plane.

Can I cascade two Sallen Key filters to get a 4th-order roll-off?
Yes. Cascading two second-order Sallen Key stages yields a 4th-order filter with an -80 dB/decade roll-off. However, you cannot just use the same component values for both stages. To maintain a proper 4th-order Butterworth response, the two stages must be tuned to different Q factors (typically Q1 = 0.541 and Q2 = 1.307). The All About Circuits Sallen-Key Guide provides excellent coefficient tables for cascading higher-order stages.