BPF

Bandpass Filter

/band-pas fil-ter/

A bandpass filter is a frequency-selective network that passes signals within a defined frequency range (the passband) while attenuating signals at frequencies above and below that range (the stopbands). The passband is bounded by lower and upper cutoff frequencies, typically defined at the -3 dB points. Bandpass filters are fundamental building blocks in receivers, transmitters, and test equipment, providing channel selection and interference rejection.

Category: Filters

Related to: Lowpass Filter, Highpass Filter, Notch Filter

Units: GHz (frequency), dB (loss/rejection)

Understanding Bandpass Filters

Bandpass filters are among the most common components in any RF system. Every receiver uses at least one to select the desired signal band and reject out-of-band interference. The performance of a BPF is characterized by its center frequency, bandwidth, insertion loss, rejection, and selectivity (how sharply it transitions from passband to stopband).

Filter Technologies

Lumped element: Uses discrete inductors and capacitors. Practical up to a few GHz. Compact but limited Q factor.
Cavity filters: Metallic resonant cavities coupled together. Very high Q (5,000+), used in base stations and satellite systems.
Waveguide filters: Iris-coupled or post-coupled waveguide cavities. Extremely low loss at mmWave frequencies.
Dielectric resonator filters: Use high-permittivity ceramic pucks. Excellent temperature stability.
Microstrip/stripline: Printed coupled-line or hairpin resonators. Low cost, easily integrated.
SAW/BAW filters: Surface or bulk acoustic wave devices. Very compact, used in mobile handsets for sub-6 GHz bands.

Key Specifications

Center frequency (f0): The geometric mean of the upper and lower cutoff frequencies.
Bandwidth (BW): The width of the passband, often specified at -3 dB points.
Insertion loss: Signal attenuation within the passband. Lower is better.
Rejection: Attenuation at specified stopband frequencies. Higher is better.
Return loss: Impedance match within the passband. Higher is better.
Group delay variation: Phase distortion across the passband. Critical for wideband digital signals.

Center frequency:
f0 = √(f_lower × f_upper)

Fractional bandwidth:
FBW = (f_upper - f_lower) / f0 × 100%

Loaded Q factor:
Q_L = f0 / BW

For a 10 GHz filter with 500 MHz bandwidth:
Q_L = 10,000 / 500 = 20

Filter Technology Comparison

Technology	Frequency Range	Typical Q	Insertion Loss	Size
Lumped element	DC - 3 GHz	50 - 200	1 - 3 dB	Very small
Cavity	0.3 - 40 GHz	2,000 - 10,000	0.1 - 0.5 dB	Large
Waveguide iris	3 - 300 GHz	5,000 - 20,000	0.05 - 0.3 dB	Large
Microstrip	1 - 100 GHz	100 - 300	1 - 4 dB	Small
SAW	0.1 - 3 GHz	500 - 2,000	1 - 3 dB	Tiny

Common Questions

Frequently Asked Questions

What is a bandpass filter used for in RF?

A bandpass filter selects a specific frequency band and rejects everything outside it. In a receiver, the BPF selects the desired signal channel. In a transmitter, it suppresses harmonics and spurious emissions. In test equipment, it isolates the signal of interest from noise and interference.

What determines the sharpness of a bandpass filter?

The filter order (number of resonator sections) determines the transition slope between passband and stopband. Higher-order filters have sharper rolloff but also higher insertion loss and more group delay variation. Chebyshev and elliptic filter responses offer sharper transitions than Butterworth designs.

What is the difference between a bandpass and notch filter?

A bandpass filter passes frequencies within a specified range and rejects those outside it. A notch (band-reject) filter does the opposite: it rejects a narrow band of frequencies while passing everything else. Notch filters are used to suppress specific interference.

Related Terms

← Balun Bandwidth →