A band stop filter is an electronic circuit that attenuates a specific, defined range of frequencies while allowing all frequencies below and above that rejected band to pass through with minimal loss. In a real circuit or installation, it changes the signal profile by surgically removing a narrow or wide interference spur—like 60 Hz mains hum or a 2.4 GHz Wi-Fi signal—without degrading the surrounding broadband data or audio. People most commonly confuse it with a band-pass filter (which does the exact opposite, keeping only the target band) or a simple low-pass filter (which blindly chops off everything above a single cutoff frequency).

Bench Rule of Thumb: If you need to kill a single annoying frequency but keep everything else intact, you need a band stop. If you need to isolate a specific frequency band and kill everything else, you need a band pass.

Real-World Band Stop Filter Applications

Before diving into circuit math, it helps to see how these filters are deployed across different engineering domains. The topology and component requirements shift drastically depending on whether you are dealing with audio signals, radio frequencies, or high-current power lines.

Application Domain Target Rejection Band Typical Filter Topology Key Performance Metric
Audio & Instrumentation 50 Hz / 60 Hz Mains Hum Active Twin-T Notch Notch Depth > 40 dB, Q=10
RF Communications 2.400 - 2.483 GHz (Wi-Fi) Cavity / SAW Band Stop Insertion Loss < 1.5 dB in passband
Biomedical (ECG/EEG) 0.1 Hz to 0.5 Hz Baseline High-Order Active BS Phase linearity in passband
Power Electronics 3rd Harmonic (150/180 Hz) Passive LC Trap Current handling > 50A RMS
Automotive Radar 76 - 81 GHz Interference Waveguide / Microstrip Rejection > 30 dB at 79 GHz

Notice the shift in topology as frequency increases. At audio frequencies, we rely on active op-amp circuits and precise RC networks. By the time we reach microwave RF applications, discrete resistors and capacitors become useless due to parasitic inductance; instead, we use Surface Acoustic Wave (SAW) devices or physically tuned cavity resonators. For a deep dive into the underlying transfer functions, the Analog Devices MT-223 tutorial provides an excellent mathematical breakdown of these topologies.

The 60 Hz Hum Killer: A Worked Numeric Example

Let's design an active Twin-T band stop (notch) filter to eliminate 60 Hz mains hum from a high-gain audio preamplifier. The Twin-T topology is favored here because it can achieve massive attenuation at the center frequency if the components are perfectly matched.

The center frequency ($f_c$) of a passive Twin-T network is defined by:

f_c = 1 / (2 * π * R * C)

Step 1: Choose the Capacitor
We need a capacitor value that yields practical resistor values. Let's select C = 270 nF (0.27 µF). Critical bench note: You must use C0G/NP0 ceramic or polypropylene film capacitors here. If you use X7R or Y5V dielectrics, their severe voltage coefficient and dielectric absorption will distort the audio signal and destroy the filter's Q-factor.

Step 2: Calculate the Resistor
Rearranging the formula to solve for R:
R = 1 / (2 * π * 60 Hz * 270e-9 F)
R = 1 / 0.000101787
R ≈ 9,824 Ω

Step 3: Select Standard E96 Values and Add Trimming
The nearest 1% standard E96 resistor is 9.76 kΩ. However, a Twin-T filter requires exact matching to achieve a deep notch. If the components are off by even 1%, your 40 dB rejection drops to 15 dB. Therefore, the professional approach is to use a 9.1 kΩ fixed 1% metal film resistor in series with a 2 kΩ multi-turn cermet trimmer potentiometer.

Step 4: The Parallel Branch (2R and C/2)
The Twin-T requires a parallel branch with twice the resistance and half the capacitance:
2R: 19.52 kΩ (Use two 9.76 kΩ 1% resistors in series).
C/2: 135 nF (Use two 270 nF C0G capacitors in series).

By buffering this passive network with a high-performance audio op-amp like the TI OPA1612 and applying a slight positive feedback loop, you can tune the Q-factor to exactly 10, yielding a razor-sharp 60 Hz rejection that leaves the 20 Hz to 20 kHz audio band completely flat. For more on active filter tuning, refer to the Electronics Tutorials band stop guide.

Band Stop vs. Band Pass: Clearing Up the Confusion

Even experienced hobbyists occasionally swap the terminology or misapply the filter type when reading a schematic. Here is the definitive breakdown of what people commonly confuse with band stop filters:

  • Band Stop vs. Band Pass: A band-pass filter is the exact inverse. It blocks low and high frequencies, allowing only a specific 'window' to pass. If you are trying to isolate a 433 MHz RF transmitter signal from background noise, you use a band-pass. If you are trying to block a 433 MHz jammer from reaching your broadband software-defined radio (SDR), you use a band stop.
  • Band Stop vs. Low-Pass/High-Pass: Low-pass and high-pass filters have a single cutoff frequency ($f_c$) and a continuous roll-off slope (e.g., -20 dB/decade). A band stop filter has two cutoff frequencies ($f_L$ and $f_H$) defining the rejected band, and the signal recovers to 0 dB attenuation outside this band.
  • Band Stop vs. Notch Filter: Technically, a notch filter is just a subset of a band stop filter. In industry parlance, 'notch' implies a very narrow rejection band with a high Quality factor (Q > 10), used for removing single frequencies like 60 Hz hum. 'Band stop' implies a wider rejected bandwidth (Q < 5), such as blocking the entire 2.4 GHz Wi-Fi spectrum.

Where You Meet This in Practice (and How to Spec One)

When moving from textbook theory to the workbench or a commercial PCB layout, you rarely build discrete LC band stop filters for RF applications anymore. Parasitic capacitance and trace inductance make discrete RF traps highly unpredictable above 100 MHz. Instead, you specify integrated modules.

Specifying RF Band Stop Modules

If you are designing a receiver front-end that needs to survive near a high-power Wi-Fi router, you will spec a surface-mount band stop filter from manufacturers like Mini-Circuits or Murata. For example, the Mini-Circuits BFCN-2500+ is a ceramic band stop filter designed to reject the 2.4 GHz ISM band. When reading the datasheet, focus on two critical parameters:

  1. Stopband Attenuation: How much does it kill the bad signal? (e.g., 35 dB minimum in the 2.4 - 2.5 GHz range).
  2. Passband Insertion Loss: How much of your good signal does it accidentally eat? A good RF band stop will have less than 1.0 dB insertion loss in the passbands (e.g., DC to 2.0 GHz and 2.8 GHz to 5.0 GHz).

The DSP vs. Analog Debate in Audio

In modern audio and biomedical instrumentation, there is a strong temptation to skip analog band stop filters entirely and rely on Digital Signal Processing (DSP). A SigmaDSP chip or an ARM Cortex-M4 can execute an IIR notch filter with perfect component matching and zero thermal drift.

However, you still need an analog band stop filter before the Analog-to-Digital Converter (ADC) if the interference is large enough to clip the ADC's input stage. If a 60 Hz mains hum is riding at 2V peak-to-peak on a 50 mV ECG signal, and your ADC reference is 3.3V, the hum will eat up your dynamic range. An analog active notch filter knocks that 2V hum down to 20 mV before it hits the ADC, allowing you to amplify the 50 mV ECG signal and digitize it with full resolution. As outlined in All About Circuits' filter chapters, understanding the signal chain's voltage limits is just as important as the frequency domain math.

Safety & Code Caveat: When designing passive LC band stop traps for power electronics (e.g., filtering 150 Hz 3rd harmonics on a 480V 3-phase line), the inductors and capacitors must be rated for the full line voltage and RMS current. Never use standard signal-grade components in mains-voltage filter traps; use UL-recognized power capacitors and ensure the trap includes a bleeder resistor to prevent lethal stored charge after de-energization.