It’s a routine, sometimes tedious mixing technique. It also can work wonders on your mix. Skillfully applied to each audio sample or virtual instrument track (and occasionally to groups of instruments), low pass and high pass filters effectively clear out sonic clutter, remove muddiness, and unmask frequencies. The process itself is fairly straightforward and doesn't require any special sound engineering skills. All you need is a set of widely available cheat sheets and some common sense to apply them.
Let's start with the theoretical basis for this method. Each acoustic or electronic musical instrument, including vocals, has its own range of effective frequencies. These are the frequencies that do the heavy lifting in terms of creating music and contributing to an instrument's unique timbre. When we record, though, an electric guitar or drums, for example, we’re capturing the entire range of the instrument—including completely unnecessary low and high frequencies.
Let's use the electric guitar as an example. Guitarists adore their instrument. It's painful for them to think that certain frequencies within its range aren't important. Sometimes they believe that barely audible overtones of the electric guitar, which can be faintly heard at higher frequencies, are something special. The same goes for the "bassiness" of some guitars and their parts.
In actual music production, however, the electric guitar operates effectively within a conditionally "incomplete" range. Nothing significant happens below 80–100 Hz, nor above 10 kHz. Experienced sound engineers know that even delicate violins don't contribute anything useful to the music above 10 kHz. They just produce unpleasant whistling and squeaking sounds there, which, generally speaking, there isn’t a huge market for.
In short, the sound of the electric guitar should be trimmed both above and below the approximate range of 80–10,000 Hz. Enjoying complex guitar overtones is best left for solo playing. If a guitarist is producing their own work, they need to try to temporarily detach themselves from being a guitar player and become a sound engineer for a bit. Then, it's time to ruthlessly sweep through the electric guitar tracks with high-pass and low-pass filters.
If you're a beginner indie musician just learning about music production, it might be hard to believe everything we’ve said so far. But you can conduct a simple experiment. Record an electric guitar part in your DAW, running it through your favorite chain of effects. Then load a parametric equalizer into the insert slot on the track.
Choose a low-pass filter curve with a steep slope—say, 30 decibels per octave, or even 36 decibels per octave. Cut off all frequencies above 80–100 Hz and listen carefully to what plays back in the guitar part (its low-frequency range). You'll hear rumble, low-end thumping, and other non-musical sounds that are entirely irrelevant to your song and only add noise.
Do the same with a high-pass filter for ultra-high frequencies above 10 kHz on your guitar track, and listen closely to what's going on there. You'll realize that nothing in this range improves your mix musically, aesthetically, or dramatically.
Apply both the low-pass and high-pass filters to your electric guitar. Graphically, the curve of your equalizer will resemble a trapezoid (remember geometry class?). The slope of the cuts is relatively safe up to values of 30–36 decibels per octave. If you cut frequencies with greater steepness, the EQ plugin may start introducing phase distortion. More accurately, it introduces these distortions in very small and almost imperceptible amounts in nearly every case (though this depends on many factors), but with extreme slopes, the distortions could become noticeable to a trained ear.
After applying the filters, you'll get a signal cleaned of low-frequency and high-frequency noise. When played solo, you might feel like the electric guitar sound is somehow incomplete, but trust us, that's an illusion. Within the context of the full mix, it'll perform much cleaner and more efficiently.
Performing similar operations makes sense for all instruments in your mix. But how do you determine the effective frequency ranges for specific instruments? Generations of sound engineers and scientists have already done this for you. You can use the power of the internet to find the effective frequency ranges of guitars, bass guitars, acoustic drums, brass and woodwind instruments, etc. Then save these cheat sheets for yourself. If necessary, print them out and hang them on the wall in your studio. For a while, you'll refer to them. After several months of practice, you'll memorize them. It's just that simple.
The primary range (fundamental frequencies) of all vocalists on Earth—both men and women—lies roughly between 100 and 1,000 Hz. Their harmonics (overtones) fall between 1,000 and 16,000 Hz. Some singers with deep basses can produce a useful musical "signal" below 100 Hz—down to about 82 Hz. Similarly, sopranos can reach above 1,000 Hz. These cases, as you might imagine, are rare and more typical in opera than pop music.
As such, you’re pretty safe cutting low and high frequencies from a vocal track—taking into account, of course, whether it's a male or female voice and the register they're singing in (and if they’re a world class opera singer or not).
Incidentally, most adult listeners have difficulty hearing frequencies above 14 kHz, if they can hear them at all. Children and teenagers can usually hear those upper harmonics around 16 kHz. However, with age—and we're not even talking about old age here, just young adulthood—most people's ability to perceive ultra-high frequencies diminishes. This applies even to professional sound engineers with trained ears. Even they are not safe from the march of time.
So, basically, paying a bunch of attention to what happens in the mix above 16 kHz is excessive and irrational. Most instruments' output in this range can safely be cut since there's no real musical activity happening there. Some producers prefer to keep cymbal vibrations in the ultra-high frequency area, arguing that it creates a subconscious sensation of "air." This claim is debatable, however, and likely reflects personal aesthetic preferences rather than objective reality.
Synthetic sounds obviously don’t exist in nature. They lack the history associated with acoustic musical instruments, and synthesizers are free from the physical constraints inherent in guitars, violins, brass instruments, and so forth.
So, what to do? And what do we use to guide our application of high-pass/low-pass filters? This is where the skill of the arranger comes into play. First, you want to strive to ensure different synth parts never (you read that right—never) play in the same register (frequency range). An exception is made only for layering IDENTICAL parts with identical notes, where synths achieve octave or double-octave doubling, or when they play in unison to create a denser, punchier sound.
Second, carefully consider what each specific synth contributes to the arrangement in terms of the drama and aesthetics of the work. Apply high-pass/low-pass filters only after the arrangement is mostly complete.
Each synth patch has timbral characteristics that define its role in the arrangement, as does the register in which a particular part is played. Simply put, if one of your synths is responsible for airy arpeggios in a high register, you can cut off all frequencies below 100–300 Hz. That's because in this low-frequency range, the specific part isn't doing anything, but the sound of the synth patch itself can bleed into it, cluttering the space where other synth parts should operate.
Often, in a mix, you only need to retain the frequency range where the individual timbre and distinctive character of the synth sound are clearly felt. Everything outside this range can usually be cut without issue.
When first working with synth music, the equalization techniques we've discussed can pose something of a challenge. This is especially true if the song's arrangement isn't well thought out, and the synths constantly step on each other's toes. In such cases, sorry, but you’re going to have to rework the arrangement.
However, if the arrangement is already clear and readable, applying high-pass/low-pass filters simply means following the logic of the arrangement itself—as in the example with the airy arpeggios.
In today's sample market, particularly in genres like EDM and hip-hop, drum samples often come heavily processed with layering and saturation. Consequently, a kick drum can roar like hell below 30 Hz and click above even 10 kHz. Similarly, snare drums can encroach upon the low-frequency range and perform excessively at ultra-high frequencies.
These samples sound extremely impressive and powerful on their own, which attracts music producers. The problem is, within the context of the entire mix, over-produced drum sounds can dominate virtually the entire effective frequency range, overshadowing tonal instruments and vocals. So, yes, high-pass/low-pass equalization should also be applied to drum samples.
If you take music production seriously enough, avoid using drum machines with purchased samples on a single stereo track of a virtual instrument. You’re not going to have sufficient control over the frequency range of synthetic and sampled drums.
You need to control the kick drum, snare, hats, cymbals, toms, and so on individually. So, create a group track called "drums" in your DAW. Assign outputs from the virtual drum machine to separate channels on the virtual mixer console, then route them so that the outputs from each drum channel go to the "drums" group bus. This way, you'll gain control over the sound of each drum element.
From there, approach equalization similar to how you would with synths. Each drum sound serves a distinct musical, aesthetic, and dramatic function. Based on the arrangement's logic, apply high-pass/low-pass filters accordingly. Remember that the kick drum essentially has no business in the range below 30 Hz—it's all unpleasant rumble and boom down there. Keep in mind that the most pleasing low-frequency "punch" of the bass drum lies in the 50–60 Hz zone. Also remember that the "click" of the kick drum shouldn't be present above 7–10 kHz.
Remember to keep the snare out of the zone below 80–100 Hz. In most cases, it interferes with the kick drum and bass lines. Similar considerations should guide you when equalizing other drum sounds.
"Apply high-pass/low-pass filters to every track in the project? To EVERY drum sample? Are you joking? Are you crazy?" a reader might think (or frantically demand answers about). We’re dead serious. If you haven’t experienced this musical magic yet, you’re going to feel its astounding effect once you apply it to at least one project with a well-thought-out arrangement.
After implementing the technique described above, your pre-mixed song will suddenly sound much clearer, brighter, and punchier, even though you haven't emphasized anything in the instruments and haven't even started compressing them. The trapezoidal (geometry!) curves of parametric equalizers actually do a huge job of cleaning up a muddy mix and preparing it for final mastering.
Once you've done this for one of your songs, you'll want to do it again. After a few months of practice, this tedious and boring task will start becoming almost automatic for you. You'll clean the frequency ranges of all instruments quickly and efficiently. And, one day, you'll look back and wonder, "How did I ever live without this?" Dead serious.
