A 2nd-order Sallen-Key low-pass filter circuit provides a -40 dB/decade rolloff using a single operational amplifier, two resistors, and two capacitors. For a 1 kHz cutoff frequency with a Butterworth response (Q = 0.707), the standard unity-gain configuration requires R1 = R2 = 11 kΩ, a feedback capacitor C1 = 10 nF, and a ground capacitor C2 = 22 nF. This topology is the benchmark for audio anti-aliasing and sensor signal conditioning when a quality factor under 10 is required.

Topology Map and the Sallen-Key Advantage

Before selecting components, you must understand the signal path. The unity-gain Sallen-Key low-pass topology relies on four specific nodes:

  • $V_{in}$: The input signal source.
  • Node 1 ($N_1$): The junction between R1 and R2. C1 bridges from this node directly to the output.
  • Node 2 ($N_2$): The junction between R2, C2 (to ground), and the non-inverting (+) input of the op-amp.
  • $V_{out}$: The op-amp output, which also ties directly to the inverting (-) input to enforce unity gain (K=1).

Why Sallen-Key Over Multiple Feedback (MFB)?

When designing an active filter, the two dominant choices are Sallen-Key and Multiple Feedback (MFB). Sallen-Key uses the op-amp in a non-inverting configuration, meaning the input impedance is high and strictly determined by R1. MFB uses an inverting topology, which heavily loads the source. Furthermore, Sallen-Key is far less sensitive to the op-amp's finite Gain-Bandwidth Product (GBW) at high frequencies. According to Analog Devices' Filter Wizard documentation, Sallen-Key is the superior choice for Q < 3 (like the Butterworth Q=0.707), while MFB is reserved for high-Q peaking filters where Sallen-Key would require unmanageable capacitor ratios.

Component Selection Matrix (E24 Values)

Theoretical filter equations yield irrational numbers. In practice, you must map these to standard E24 resistor and E12 capacitor values. The table below provides pre-calculated, real-world component values for a unity-gain Butterworth Sallen-Key filter across four common audio and sensor cutoff frequencies. These values assume a dual-supply op-amp (e.g., ±12V or ±15V).

Target $f_c$ R1 & R2 (E24) C1 (Feedback, E12) C2 (Ground, E12) Min. Op-Amp GBW Recommended Op-Amp
100 Hz 110 kΩ 10 nF 22 nF > 10 kHz TL072 / LM358
1 kHz 11 kΩ 10 nF 22 nF > 100 kHz TL072 / NE5532
10 kHz 1.1 kΩ 10 nF 22 nF > 1 MHz NE5532 / OPA2134
100 kHz 110 Ω 10 nF 22 nF > 10 MHz OPA2134 / AD823
Dielectric Warning: Never use X7R or Y5V ceramic capacitors for C1 and C2 in audio or precision DC paths. X7R ceramics exhibit severe capacitance drop under DC bias and introduce piezoelectric microphonics (thump noise). Use C0G/NP0 ceramics or polyester film capacitors (like the WIMA MKS series) to maintain your target Q and cutoff frequency.

Design Walkthrough: 1 kHz Butterworth Anti-Aliasing Filter

Let's trace the math and physical selection for the 1 kHz design to understand how the E24 values were derived. For a unity-gain Sallen-Key filter, the cutoff frequency ($f_c$) and Quality Factor ($Q$) are defined by:

$f_c = \frac{1}{2\pi\sqrt{R_1 R_2 C_1 C_2}}$
$Q = \frac{\sqrt{R_1 R_2 C_1 C_2}}{C_1(R_1 + R_2)}$

To achieve a Butterworth response ($Q = 0.707$ or $1/\sqrt{2}$) with $R_1 = R_2 = R$, the capacitor ratio must be exactly $C_2 = 2 \times C_1$.

  1. Pick C1: Start with a standard 10 nF film capacitor.
  2. Calculate C2: $2 \times 10\text{ nF} = 20\text{ nF}$. The nearest standard E12 value is 22 nF. This shifts Q slightly to 0.738, which is virtually indistinguishable from 0.707 in real-world audio applications.
  3. Calculate R: Plugging $f_c = 1000$, $C_1 = 10\text{ nF}$, and $C_2 = 22\text{ nF}$ into the frequency equation yields an ideal resistance of 10,693 Ω.
  4. Pick R1 & R2: The nearest E24 value is 11 kΩ. Using 11 kΩ shifts the actual cutoff frequency to roughly 970 Hz, well within the 5% tolerance of the resistors.

For the active element, the TL072 op-amp is an excellent bench choice. Its JFET inputs provide low bias current, and its 3 MHz GBW is thirty times higher than our 100 kHz minimum requirement, ensuring the op-amp's internal phase shift won't degrade the filter's stopband performance.

Breadboard Verification and Extreme Failure Modes

Simulations assume ideal grounds and perfect power rails. Breadboarding reveals the physical reality of parasitic inductance and power supply noise. Follow this exact sequence to verify your build.

Step-by-Step Breadboard Testing

  1. Rail Decoupling: Wire your ±12V power rails. Place a 10 µF electrolytic and a 100 nF MLCC ceramic capacitor across the power pins of the TL072 (Pins 8 and 4) as close to the IC body as physically possible. Skip this, and your filter will oscillate at high frequencies.
  2. Op-Amp Feedback: Wire Pin 1 (Output A) directly to Pin 2 (Inverting Input A) to establish the unity-gain buffer before adding the passive network.
  3. Passive Network: Insert R1, R2, C1, and C2 according to the node map. Keep the leads of C1 and C2 short to minimize stray inductance.
  4. Signal Injection: Connect a function generator to $V_{in}$. Set it to a 1 Vpp sine wave with a 0V DC offset.
  5. Sweep and Measure: Connect an oscilloscope to $V_{out}$. Sweep the generator from 10 Hz to 10 kHz. At 1 kHz, your scope should read exactly 0.707 Vpp (the -3 dB point). At 10 kHz (one decade up), it should read roughly 0.01 Vpp, confirming the -40 dB/decade rolloff.

Behavior and Failure Mode Matrix

When a circuit fails on the bench, you need to know what a specific component fault looks like at the output. Here is the failure-mode contrast for the Sallen-Key topology:

Component Failure Mode Circuit Result at $V_{out}$ Diagnostic Measurement
C1 (Feedback) Open Rolloff degrades from -40 dB/dec to -20 dB/dec. Acts as a 1st-order passive RC filter. Measure 0.316 Vpp at 10 kHz instead of 0.01 Vpp.
C1 (Feedback) Short $N_1$ ties directly to $V_{out}$. R1 and R2 form a simple voltage divider with C2. Severe high-frequency noise. AC signal passes with minimal attenuation; DC offset shifts.
C2 (Ground) Short $N_2$ is hard-grounded. Op-amp non-inverting input sits at 0V. Output locks to 0V DC. Read 0.00V on scope regardless of $V_{in}$ frequency.
R2 Open $N_2$ floats. Op-amp (+) input picks up stray 50/60Hz mains hum and RF noise. Output rails to +12V or -12V, or exhibits wild low-frequency oscillation.
Op-Amp Feedback Open (Pin 1 to 2) Op-amp operates in open-loop as a comparator. Output saturates to the positive or negative rail. Read steady +11V or -11V DC (depending on input offset voltage).

Understanding these failure states transforms troubleshooting from a guessing game into a systematic elimination process. If your scope shows a -20 dB/decade rolloff instead of -40 dB/decade, you immediately know C1 has failed open or the breadboard contact at the $V_{out}$ feedback node is loose. By anchoring your design in exact E-series values and verifying the physical node behavior, the Sallen-Key filter becomes a highly predictable, robust building block for any mixed-signal workbench.