A band stop filter circuit (often called a notch or band-reject filter) attenuates a specific, narrow band of frequencies while allowing signals above and below that band to pass unimpeded. If you are designing an audio preamp, an ECG sensor, or a high-gain instrumentation amplifier, 60Hz (or 50Hz) mains hum is the most common enemy. The direct answer for rejecting this low-frequency interference without introducing active op-amp noise is the passive Twin-T RC notch filter.

Unlike simple RC low-pass or high-pass rollovers, a band stop filter circuit relies on destructive interference. It splits the input signal into two paths—a low-pass network and a high-pass network—and recombines them out of phase at the target frequency, effectively canceling it out. Below is a complete bench-to-breadboard guide for designing a 60Hz Twin-T filter using real, purchasable component values.

Twin-T Topology and Node Mapping

The Twin-T topology consists of two distinct T-shaped RC networks wired in parallel. One T-network acts as a low-pass filter (two resistors and a shunt capacitor), and the other acts as a high-pass filter (two capacitors and a shunt resistor).

To troubleshoot or simulate this circuit, you must understand the specific node labels:

  • Vin (Input Node): Where the raw signal enters. Splits immediately into the R1/C1 junction and the C1/R1 junction.
  • Vout (Output Node): Where the recombined signal exits. Formed by the junction of R2 and C2.
  • Node A (Low-Pass Summing Node): The junction between R1, R2, and the shunt capacitor C3. This node routes high frequencies to ground.
  • Node B (High-Pass Summing Node): The junction between C1, C2, and the shunt resistor R3. This node routes low frequencies to ground.
  • GND (Common Ground): The shared reference for C3 and R3.
Bench Tip: The symmetry of this circuit is absolute. If the ratios between the R and C branches are off by even 2%, the destructive interference fails, and your "notch" becomes a shallow dip rather than a deep null.

Component Selection and Behavioral Matrix

Let’s design a band stop filter circuit targeting exactly 60Hz. The center frequency ($f_c$) formula for a symmetrical Twin-T network is:

$f_c = \frac{1}{2 \pi R C}$

We need to pick standard values. Let’s set $C = 100\text{nF}$ (0.1µF). Solving for R yields $26.52\text{k}\Omega$. We will use 26.7kΩ 1% metal film resistors (a standard E96 series value), which shifts our actual center frequency to 59.6Hz—close enough to nail the 60Hz fundamental and its immediate sidebands.

In a standard Twin-T, the shunt components must be exactly half/double the series components: $R3 = R/2$ and $C3 = 2C$.

Component Ref Role in Topology Calculated Value Real-World BOM Selection Tolerance / Dielectric
R1, R2 Series Low-Pass 26.52 kΩ 26.7 kΩ (Vishay MFR-25) 1% Metal Film
R3 Shunt High-Pass 13.26 kΩ 13.3 kΩ (or two 26.7k in parallel) 1% Metal Film
C1, C2 Series High-Pass 100 nF 100 nF (Kemet C0G) 5% C0G/NP0
C3 Shunt Low-Pass 200 nF 2x 100nF in parallel 5% C0G/NP0

Element Drift Behavior Matrix

When debugging a physical breadboard, components drift due to temperature, parasitic breadboard capacitance, or poor solder joints. Here is how the circuit behaves when specific elements deviate from their ideal state.

Element Changed Direction of Change Effect on Center Frequency ($f_c$) Effect on Notch Depth (Q-Factor)
R1 or R2 Increases (e.g., +5%) Shifts lower (asymmetric split) Severe degradation (notch fills in)
C1 or C2 Decreases (e.g., -10%) Shifts higher Moderate degradation
R3 (Shunt) Increases (Open circuit) Loss of high-pass ground path Notch disappears; becomes low-pass
C3 (Shunt) Increases (e.g., +20%) Shifts lower Asymmetric notch skirt (steeper high-side)

Why Twin-T Over LC or State-Variable Topologies?

When searching for a band stop filter circuit, you will encounter LC (inductor-capacitor) tanks and active State-Variable op-amp topologies. Here is why the passive RC Twin-T wins for low-frequency mains hum rejection:

  • vs. LC Tank Filters: To notch 60Hz with an LC circuit, you need a massive inductor (often >1 Henry). Large inductors have high DC resistance (insertion loss) and, critically, act as magnetic antennas. An inductor designed to block 60Hz hum will physically absorb 60Hz magnetic flux from nearby power transformers, injecting the very hum you are trying to eliminate. The Twin-T uses only resistors and capacitors, making it immune to magnetic interference.
  • vs. Active State-Variable Filters: A state-variable filter requires three op-amps and precise matching to create a notch. While it offers a tunable Q-factor, it introduces op-amp voltage noise, requires dual power rails (±12V or ±15V), and consumes board space. The passive Twin-T requires no power supply, generates zero thermal noise beyond the Johnson-Nyquist noise of its resistors, and can be inserted directly into high-impedance signal paths.

The primary drawback of the passive Twin-T is its inherently low Q-factor (maximum theoretical Q is ~0.33). The notch is relatively wide. If you need a razor-thin notch (Q > 10), you must add an op-amp "bootstrap" loop to feed a portion of the Vout signal back into the GND node of R3/C3, effectively increasing the Q without altering the center frequency (All About Circuits).

Extreme Failure Modes: Opens and Shorts

Understanding what happens when a component catastrophically fails is crucial for designing fault-tolerant sensor front-ends. Unlike simple series/parallel circuits, the dual-path nature of the Twin-T creates distinct failure signatures.

Open Circuit Failures

  • R1 Opens: The low-pass path to Node A is broken. The circuit loses its ability to pass low frequencies. Vout will only show high-frequency content. The band stop filter circuit effectively degrades into a first-order high-pass filter governed by C1 and the load impedance.
  • C3 Opens: The shunt path to ground for high frequencies is removed. The low-pass T-network can no longer divert high frequencies to ground. The notch depth will collapse entirely, and the circuit will pass all frequencies with slight phase shifting.

Short Circuit Failures

  • C1 Shorts: The high-pass path becomes a direct wire. Low frequencies bypass the C1/C2 network and flood directly into Vout. The notch is destroyed, and the circuit acts as a first-order low-pass filter (governed by R1/R2 and C3).
  • R3 Shorts: Node B is hard-tied to ground. C1 and C2 now act as AC coupling capacitors directly to ground, severely attenuating the entire signal spectrum. Vout will drop to near zero millivolts across all frequencies.
Capacitor Dielectric Warning: Never use X7R, Y5V, or Z5U dielectrics for C1, C2, or C3 in a low-frequency notch filter. These Class II/III ceramics exhibit severe capacitance loss under DC bias and generate piezoelectric "microphonic" noise when subjected to mechanical vibration. Always specify C0G (NP0) dielectrics, or use polypropylene film capacitors (like WIMA FKP2) if the required capacitance exceeds 470nF (Electronics Tutorials).

Step-by-Step Breadboard Verification

Do not trust SPICE simulations blindly; breadboard parasitic capacitance (typically 2pF to 5pF per node) can skew high-frequency filters, though at 60Hz, the 100nF capacitors easily dominate. Follow this exact test sequence to verify your physical build.

  1. Pre-Flight DMM Check: Before applying power, use a multimeter in resistance mode. Measure across Vin and GND. You should read roughly 53.4kΩ (R1 + R2 in series, parallel with the R3 branch). Measure Vout to GND; it should read open-loop (OL) or very high impedance, confirming no accidental solder bridges to ground.
  2. Signal Injection: Connect a function generator to Vin. Set it to output a 1Vpp (peak-to-peak) sine wave with a 0V DC offset. Terminate the generator with a 50Ω coaxial cable, but do not place a 50Ω termination resistor at the filter input; the Twin-T expects a low-impedance drive source.
  3. Baseline Passband Measurement: Connect your oscilloscope probe (set to 10x attenuation, compensated) to Vout. Set the function generator to 10Hz. Measure the Vout amplitude. It should be roughly 0.9Vpp to 1.0Vpp (minimal insertion loss). Repeat at 2kHz to verify the high-frequency passband.
  4. Sweeping the Notch: Slowly sweep the function generator frequency from 40Hz up to 80Hz. Watch the oscilloscope amplitude. You will see the waveform shrink dramatically as you approach 59.6Hz.
  5. Null Depth Verification: Stop exactly at the minimum amplitude point. Switch the oscilloscope to measure RMS voltage or use the scope's FFT (Fast Fourier Transform) math function. A well-built passive Twin-T with 1% components should yield a notch depth of -25dB to -35dB relative to the passband. If your notch is only -10dB, your resistor ratio (R1/R2 to R3) is mismatched. Swap R3 for a multi-turn 25kΩ trimmer potentiometer and adjust while monitoring the scope to dial in a perfect null.

By strictly adhering to 1% resistor tolerances and C0G capacitor dielectrics, this band stop filter circuit will reliably strip 60Hz interference from sensitive analog signals without the noise penalties of active filtering. If your application demands a tighter notch width, buffer the Vout node with a unity-gain op-amp (like an OPA1612) and feed a voltage divider back to the R3/C3 ground node to artificially boost the Q-factor.