A low-pass filter lets low frequencies through and blocks high frequencies. A high-pass filter does the opposite, letting high frequencies through while blocking low ones. Both work by defining a cutoff frequency: the specific point where the filter starts doing its job. These two filter types show up everywhere, from the speakers in your car to the power supply in your laptop to the photos on your phone.
How Each Filter Works
Every signal, whether it’s sound, electrical current, or image data, is made up of different frequencies. Some change slowly (low frequency), and some change rapidly (high frequency). Filters sort these frequencies by allowing some to pass and reducing the strength of others.
A low-pass filter permits everything below its cutoff frequency to pass through freely. Frequencies above that cutoff get progressively weakened. Think of it as a gate that only opens for slow, steady signals. A rumbling bass note passes through easily; a piercing cymbal crash gets blocked.
A high-pass filter does the reverse. It permits everything above its cutoff frequency and weakens everything below it. The cymbal crash sails through; the bass rumble gets cut.
The cutoff frequency itself is defined as the point where the output signal drops to about 70.7% of the input signal’s strength. Engineers call this the “3 dB down point” because that percentage translates to a 3-decibel reduction in power. It’s the standard marker for where a filter transitions from passing a frequency to attenuating it.
Filter Order and Sharpness
No filter flips from “fully passing” to “fully blocking” at the exact cutoff frequency. Instead, there’s a gradual slope called the roll-off. How steep that slope is depends on the filter’s order.
A first-order filter (the simplest design) rolls off at 6 dB per octave. That means every time the frequency doubles past the cutoff, the signal drops by 6 dB. A second-order filter doubles the steepness to 12 dB per octave. A fourth-order filter reaches 24 dB per octave. The pattern is straightforward: multiply the filter order by 6 to get the roll-off rate in dB per octave. Higher-order filters separate wanted and unwanted frequencies more cleanly, but they require more components and more careful design.
Passive vs. Active Filters
Filters come in two broad categories based on the components they use.
Passive Filters
Passive filters use only resistors, capacitors, and sometimes inductors. They need no external power supply, which makes them simple and reliable. The simplest low-pass filter is just a resistor and a capacitor arranged like a voltage divider, with the capacitor replacing the second resistor. Swap their positions and you get a high-pass filter instead.
Passive filters handle high frequencies well since they aren’t limited by the speed of any amplifier chip. They also generate very little electrical noise, producing only the thermal noise from their resistors. That makes them a good fit for sensitive applications and for circuits carrying large voltages or currents. The downsides: they can’t amplify a signal (only weaken unwanted parts), and any design beyond the most basic often requires inductors. Precision inductors are expensive, hard to source in exact values, and physically bulky, which makes complex passive filters costly to manufacture at scale.
Active Filters
Active filters add an amplifier (typically an op-amp) to the mix, paired with resistors and capacitors in a feedback loop. This eliminates the need for inductors entirely, which is a significant practical advantage. Active filters can also amplify the signal as they filter it, and they offer high input impedance and low output impedance, meaning they play nicely with whatever circuit comes before or after them.
The tradeoff is that active filters need a power supply to run the amplifier, and their performance at very high frequencies is limited by how fast the op-amp can respond. For most audio and mid-range electronic applications, though, active filters are easier to design and more cost-effective than their passive counterparts.
Audio and Speaker Crossovers
One of the most familiar uses of these filters is inside speaker systems. Different speakers are physically designed to handle different frequency ranges: subwoofers for bass, midrange drivers for vocals and instruments, and tweeters for the highest frequencies like cymbals and harmonics. Filters divide the audio signal so each speaker only receives the frequencies it can reproduce well.
A low-pass filter set around 60 to 80 Hz sends only deep bass to the subwoofer. A high-pass filter set around 3,000 Hz sends only the highest frequencies to the tweeter, protecting it from the large vibrations that low-frequency signals would cause. The midrange driver typically gets a band-pass filter (essentially a low-pass and high-pass filter combined) that passes frequencies between roughly 80 Hz and 3,000 Hz. These are common starting points; specific systems may benefit from slightly different crossover frequencies depending on the speakers and the listening environment.
Power Supplies and Clean Electricity
Every device that converts AC wall power into the DC power your electronics need faces a problem: leftover ripple. The conversion process leaves small, rapid voltage fluctuations riding on top of the smooth DC output. These fluctuations are high-frequency noise, and they can cause audible hum in audio equipment or errors in sensitive digital circuits.
Low-pass filters solve this by letting the steady DC voltage (zero frequency) pass through while blocking the high-frequency ripple. In practice, power supplies often use multiple filtering stages, combining inductors (often ferrite beads) and capacitors to push ripple below 1 millivolt. This is why you’ll find clusters of capacitors near the power input on almost every circuit board: they’re the low-pass filter that keeps the power clean.
Image Processing
Filters aren’t limited to electrical signals. In digital image processing, the same concepts apply to spatial data. Instead of time-based frequencies, images have spatial frequencies: how quickly pixel values change from one spot to the next. A smooth sky has low spatial frequency. A sharp edge between a dark building and a bright sky has high spatial frequency.
A low-pass filter applied to an image smooths it out by averaging neighboring pixel values together. This blurs fine details but effectively reduces noise, which is useful for cleaning up grainy photos or simplifying an image before further analysis. A high-pass filter does the opposite: it replaces each pixel’s value with the difference between it and its neighbors’ average. This highlights edges, lines, and sudden transitions, making it a core tool in edge detection and image sharpening.
Combining both is common. A photographer’s “sharpen” tool, for instance, often works by extracting the high-frequency detail from an image using a high-pass filter and then adding a portion of that detail back to the original. This enhances edges without amplifying noise as aggressively as a pure high-pass filter would.
Other Common Applications
- Radio and communications: Low-pass filters remove unwanted harmonics from transmitted signals, keeping broadcasts within their assigned frequency band. High-pass filters reject interference from lower-frequency sources.
- Audio recording: High-pass filters (sometimes labeled “low cut” on microphones and mixing boards) remove rumble from wind, air conditioning, and handling noise, typically cutting everything below 80 to 100 Hz.
- Sensor data: Accelerometers, temperature sensors, and other instruments produce noisy raw data. A low-pass filter smooths out the rapid fluctuations so the meaningful, slower-changing signal is easier to read.
- Equalizers: The bass and treble knobs on a stereo are simplified filter controls. Turning up the bass boosts the output of a low-pass filter region; turning up the treble boosts the high-pass region.
At their core, high-pass and low-pass filters are the same idea applied in opposite directions. Once you understand that every signal is a mix of frequencies and that these filters simply select which part of that mix to keep, you can spot them at work in nearly every piece of technology you use.