Music Production 101: EQ

Equalization, better known as simply “EQ”—you suspect the pros are using this all the time and you’ve got to insert an EQ plug-in on your tracks to start shaping the sound.

But what are you doing when you turn those knobs or pull those little nodes down, and what is this mysterious value called “Q” supposed to be?

You can throw an EQ on one of your tracks, move some sliders or knobs, and hear the quality of the sound change. Maybe it gets a little more bright and brittle sounding, or dark and muddy.

Should you gently ease regions of the EQ curve down, or just crank them up until it sounds good?

I hope to answer these questions and more, while making this very fundamental tool second nature for you.

Get Your ‘Freq’ On

One of the first things you’ll notice on an EQ is the range of frequencies. A full range EQ will display a scale of frequencies starting from 20 Hz and ending at 20,000 Hz (or 20kHz).

flat
Figure 1. An EQ typically allows control of frequencies between 20 to 20k Hz.

This 20 to 20k Hz range is significant because this is the standard range of frequencies that a human being can hear.


study-1355437_1280Nerd Note

Hz (“Hertz”) is a value arising from the amazing world of physics that generally means “cycles per-second”! A cycle refers to one oscillation (or vibration, if you will, since that’s what sound is). A sound with a frequency centered at around 6,000 Hz vibrates about 6,000 times within a single second.


Ranges

Sounds on the lowest end of the frequency range, about 20-60 Hz, can be considered “sub-bass“. We can kinda hear sub-bass through our monitors or speakers if we listen closely, but these frequencies are more so felt and are best projected by sub-woofers (like the kind that shake the earth in your car when you are bumping that mixtape!).

A little higher in the frequency range, from about 60-250 Hz, is the familiar “bass” range where we can clearly hear the low-end of a sound. Instruments like kick drums and bass guitars or bass synths have characteristic frequencies in this range.

Low-mids are usually said to be in the 250-500 Hz range, Mids in the 500-2,000 Hz Range, and Upper-mids in the 2,000-7,000 Hz region.

The highs occupy the range from roughly 7,000 Hz and up.

Frequencies over about 12,000 Hz are increasingly hard to hear, and many producers and mixers refer to these upper-high values by nebulous terms like “brilliance”, “air” or “shimmer”.

Every sound in your music is a composite of different frequencies (unless you make avant-garde ambient music with pure sine waves or something).

Understanding that any sound you hear is just a distribution of different frequencies is all it really takes to understanding the concept of EQ, but navigating and manipulating the different settings on an EQ, and knowing when to use the EQ in the first place, are the aspects that take practice and hands-on work.

Thankfully, EQ is a a lot more intuitive than something like Compression!

OK, So When Should I Use EQ?

Sometimes simply lowering the volume of an instrument is all it takes to balance a mix. A subtle fader move could really be all that is needed.

Most other times, important frequencies in one instrument interfere with the characteristic frequencies in another instrument, and both instruments wind up suffering in terms of loudness and separation. This unwanted effect is technically called frequency masking.

The primary purpose of EQ in mixing is to help instruments get along.

When certain instruments have overlapping frequencies, they can smear together, obscure one another, and generally just fight one another to be heard. This creates a lot of unpleasant noise, texture or volume fluctuations.

With EQ, you can trim off unwanted frequencies that really contribute nothing to the perception of an instrument. As an example, we don’t really need the lowest resonance of a guitar to know that we’re hearing a guitar, so we could knock out some of the lowest frequencies and free up some space for the kick, toms or bass. Gotta be careful with this, because too much low resonance taken off and your heavy guitars start to sound thin!

The other big job for the EQ is to sculpt the tonality of an instrument, for instance brightening up a dull-sounding cymbal a bit, or easing down an irritating “nasally” sound in a vocal recorded on a cheap mic.

You don’t need any kind of fancy EQ to do these jobs, you just need to know what you’re looking at with a basic EQ in front of you, like the stock EQ’s in your DAW.

‘Curves’ vs ‘Dials-Only’

All DAW’s come with EQ’s of their own, and these are perfectly good for getting the job done. They all look a little different, but achieve the same end result.

Logic’s native Channel EQ (Figure 1) is an example of an easy-to-read type of EQ called a multiband EQ. This particular EQ shows a graph-like display with adjustable ‘curves’.

Clicking on Channel EQ’s Analyzer button shows a real-time frequency distribution curve. I said above that sounds are just “distributions” of frequencies, so you can see how representing them in such a form would be useful to visualize the sound.

Logic's Channel EQ w Analyzer
Figure 2. I like my EQ’s like I like my women—curvy. Logic’s native Channel EQ is my go-to plug-in. Easy to see at a glance how frequencies are distributed for a given sound at any given moment thanks to the  analyzer.

Other EQ’s, like Stillwell Audio’s awesome Vibe EQ, show some dials and no “curves” display. These can feel a bit less intuitive to beginners than something like Channel EQ.

VibeEQ
Figure 3. Stillwell Audio’s Vibe EQ is a ‘dials only’ sort of EQ that also adds a little bit of color.

Even as long as I’ve been at this stuff, I still tend to use ‘curves’ EQ’s as my first go-to, but you can get some interesting results with a ‘dials-only’ EQ, as it forces you to use only your ears.

Both types of EQ are classified as parametric EQ‘s, by the way!

The Mysterious Q

If you’re looking at an EQ plug-in, you’ll see basic parameters like gain and frequency that are pretty obvious. But then there is Q …

There is a lengthy, jargon-saturated version of what is that would thrill an engineer or physicist, and then there is the everyday concept that helps us get down to business, which we’ll jump into now!

‘Q’ stands for ‘quotient of change’ or ‘quality factor‘ depending on who you ask. Just refer to our friendly Channel EQ for a visual representation of the Q-factor.

Below you will see a frequency band with a low Q (0.30) and a high Q (100). What do you notice about the shapes of the corresponding curves?

EQ-lowQ
Figure 4. a bell curve with a low Q-factor setting of 0.30.
EQ-highQ
Figure 5. a bell curve with a very high Q-factor setting of 100.

It looks like a higher Q setting narrows that particular band’s width, right? That’s the exact terminology used by engineers—bandwidth.

Lowering the Q value has the effect of increasing a band’s width.

The width of the band tells us how many of the surrounding frequencies either side of our central frequency are controlled by that EQ band.

You’ll notice that a band has gentle slopes on the left and right of its peak, this allows more natural sounding changes versus having a hard boundary between which frequencies are changed and which are not. Having hard boundaries would produce an unnatural sound when shaping the sound with EQ.

Changing the Q has the effect of not only widening the band but also making the slopes more gentle or more steep.

Really narrow/steep bands, from a high Q setting, are useful for sharp-shooting problematic frequency regions in mixing and mastering stages, which I’ll get into in the near future.

Filtering: ‘Cut’ -or- ‘Pass’

Many EQ’s have either preset curves or dedicated knobs for filtering.

You’ll see the words Low-Cut, or High-Cut. Some EQ interfaces, like the stock EQ’s in Pro Tools, use the respective terms High-Pass Filter (HPF) and Low-Pass Filter (LPF), instead. These terms are interchangeable.

Here is what a Low-Cut aka High-Pass Filter looks like applied to Channel EQ.

low-cut-high-pass.png
Figure 6. A Low-Cut aka High-Pass filter in action, rolling-off frequencies ~35.5 Hz and below.

The primary frequency in the picture above is 35.5Hz, meaning that frequencies lower than that are filtered out, or rolled-off, to use a common audio engineering term. How steeply or gradually they are tapered off depends on the Q-factor.

We can do the same sort of filtering or rolling off on the opposite end of the spectrum. Below, a high-cut aka low-pass filter is engaged on Channel EQ, cutting out frequencies that are higher than about 6,000 Hz, and letting those lower than 6,000 Hz pass through.

Screen Shot 2018-05-29 at 5.33.01 AM
Figure 7. A High-Cut aka Low-Pass filter in action, snipping off frequencies 6,000 Hz and higher.

Shelving

A shelf is kind of the opposite of a cut; instead of filtering out frequencies beyond the set frequency, a shelf will boost all frequencies beyond a set point by a given amount.

A shelf doesn’t maximally affect the volume the way a cut does. As you can see in the image below, the high-shelf applied to this track gently adds up to 5.5 dB of boost to all frequencies higher than 4,550 Hz.

You can also pull the shelf down to reduce the volume of a frequency range down to a set dB level.

The Q setting of a shelf will specify how steep or gradual the boost is over the frequency range.

high shelf
Figure 8. A +5.5 dB high-shelf is applied to the frequencies at about 4,550 Hz and up. Such an EQ move can bring out the ‘shimmer’ or ‘air’ in a dull sounding drum cymbals track, for instance.

To Cut or to Boost?

Listen to this portion of a trip-hop/world track I wrote and produced under my Keith Kaspian handle, called ‘Divinations’.

 

I layered in an acoustic, gain-saturated drum kit over the synth drums for more impact and excitement. And while it sounds interesting, it doesn’t have the “bite” or “edge” I’m looking for.

So I decided to do a gentle, wide (Q=0.46) boost centered at 3000 Hz, where I found that some of the “excitement” lives. It looks like this:

EQ-boost-divinations
Figure 9. I applied a gentle (Q=0.46) +3dB boost to bring out the edginess and excitement in a drum layer.

The boost gives me the “bite” and “edge” I wanted, as you can hear:

 

I tend to boost and cut with a low Q setting (wider bandwidth) when I am just trying to get instruments to get along in the mix. For surgical cuts, I use a much more narrow bandwidth (higher Q setting).

Whether you should cut or boost really depends on what you want to or need to achieve to get the mix sounding better.

Maybe there is an overly-bright ringing in your snare drum that is driving you nuts; in this case you could do a gentle cut somewhere in the upper-mids or low-highs to bring that ringing sound down below the overall level of the mix.

Perhaps your bass guitar needs a little thunder, in which case a gentle boost around 125-175 Hz might be the answer.

Sometimes, an instrument track is fine as is and the best mixing decision you can make is to leave it alone!

Cheater, Cheater!

There are lots of references out there that give frequency ranges and which instrument characteristics they affect (e.g. “250-300 Hz is the ‘muddy’ region of a bass guitar”). These are useful and often have nice cheat-sheet charts for when you’re starting out.

Cheat-sheets can become like training wheels, preventing you from learning to ride the bike for real.

Instead of relying too much on cheat-sheets, I would advise you, as the Cobra Kai dude did Daniel-san, to instead learn to sweep.

Sweeping

Performing a frequency sweep to hunt for specific characteristics in a sound is a technique you should really get comfortable with.

To perform a sweep on a ‘curves’ EQ, simply adjust the Q of any of the the middle, bell-shaped bands to something like 50-80 values (don’t use the cut or shelves for this). Grab the node that corresponds to that band and pull it up to between +10 and +15 dB. Sweep the node from left to right along the frequency range and listen for which components of the sound start to jump out.

For a ‘dials only’ EQ, like Vibe EQ or Pro Tools’ EQ, it’s not too hard—just take a knob, for example the MF or ‘mid-frequency’ knob. Below it you will usually see a corresponding Gain knob. Increase the Gain value to a little more than halfway between the neutral and maximum values. Now, go to the MF knob and rotate clockwise and counter-clockwise to perform a sweep in the middle-frequency range.

When you nail down the characteristic you’re looking for (e.g. the ‘click’ of a kick drum’s beater really start to jump out), leave the frequency adjustment alone and simply ease down the gain knob to the desired level.

I’ll be doing a dedicated Tech Tip post to sweeping in the not-too-distant-future, with some video examples, so stay tuned!

EQ as FX

The filtering capability of an EQ makes it another useful toy in creative sound design and experimentation.

A very common vocal effect is the telephone or broken little speaker effect that can add sultriness, mystery and intrigue to the voice. This is achieved by very hard low-pass and high-pass filters working together to reduce the frequency range of the vocal to mostly the mids (adjusted to taste).

EQ Phone Filter
Figure 10. A low-cut and a high-cut filter working together to achieve the mid-range-only telephone effect heard on many vocal tracks in many productions.

I also like to use this lo-fi sounding effect on an entire drum track just before a song section kicks in at full impact, for dramatic effect.

EQ Can’t Perform Miracles

I made the mistake early on thinking that I could ‘fix things later with EQ’.

Often times, I’d have an OK guitar tone going into my DAW for recording, although maybe there was quite a bit of hissing high-frequencies making it less than ideal. I would think “hey, that’s what they invented EQ for, right!?”

Wrong.

Garbage in, garbage outor GIGO, is an audio engineer’s mantra.

EQ is a tool for balancing the way sounds, instruments and tracks fit in with one another.

You can’t expect EQ to enhance the quality of a horrible sounding recording. You may be able to improve it a little, but getting it right at the source is critical for better productions.

EQ – Always Got Your Back

EQ isn’t just a mixing thing. You’ll find a use for EQ during every major step of music production, even before recording during the creative composing stage!

For instance, I have EQ’s armed on my channel strips every time I record guitar so that I am already using a low-cut (or high-pass) filter to filter out a lot of the rumbling mess around 80 Hz and below that is not essential for guitars. Such frequencies contribute nothing to the sound of a guitar and just eat up energy in your mix.

During mastering, some very subtle EQ’ing can help to give the final polish on a song and to also make sure the songs constituting an album all share a similar sonic vibe.

If you can grasp the essence of the ever-present and powerful EQ, you’ll be able to intelligently fiddle around with any EQ put in front of you.

There are many types of EQ’s out there with varying interfaces and nuts-n-bolts under the hood, but a general knowledge of how sounds are just composites of frequencies goes a long way to becoming capable with EQ’ing.


Let me know if this has been a useful article! Go on and share this post, and feel free to post questions in the comments!

2000px-F_icon.svg.png  Like and Follow on Facebook!

7 Comments Add yours

  1. rowan says:

    I learned something new. Thanks.

    Like

    1. Keith Prater says:

      Thanks for reading, Rowan! I’m glad it proved useful for you.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s