Tools for Mixing: EQ, part 2
EQ types: high pass and low pass filters
You’re probably very familiar with the simplest type of EQ:
Fig. 9. Though an electric guitar’s tone knobs can be wired to apply many different types of EQ to the sound of a guitar, at its most basic, turning the knob applies a low pass filter to the sound, gradually lowering the level of harmonics at the higher end of the frequency spectrum.
Here’s a low E on an electric guitar, with the tone knob at its brightest setting. This allows all the frequencies to pass through without reduction:
Crank the tone knob down, and higher frequencies are blocked — or rolled off — while lower frequencies are allowed to pass through; hence the name, low pass filter:
Just as the low pass filter attenuates (reduces) high frequencies and allows low frequencies to pass through, there is another EQ type that rolls off (reduces) low frequencies while allowing high frequencies to pass: the high pass filter. It’s not just guitars that utilize this simple EQ type. When you’re mixing, usually you’ll use high pass and low pass filters that are built into each channel of your mixer, which allow you to set the frequency at which the attenuation begins (also called the cutoff frequency).
Although their structure may be simple, low pass filters can be very effective at quickly providing solutions such as:
- Removing hiss in a particular track
- Isolating and emphasizing a low-frequency sound, such as a kick drum
Similarly, high pass filters can quickly give you remedies such as:
- Eliminating rumble, mic stand bumps, or low frequency hum on tracks that primarily have high-frequency energy in the music
Here’s how Reason’s low pass and high pass filters work.
Fig. 10. Reason provides a low pass filter (LPF) and a high pass filter (HPF) on each mixer channel, at the very top of the EQ section. You engage them simply by clicking their respective “on” buttons, and then you can set the frequency above or below which the attenuation begins.
Fig. 11. Reason’s low pass filter lets you set the rolloff point from 100 Hz up to 20 kHz. The angle of the roll-off has particular terminology and characteristics: The angle itself is called the “cutoff” or “knee,” and the “slope” or “curve” reduces the signal by 12 dB per octave. Here you can see how much is allowed to pass at the lowest setting (yellow area only) and how much passes at the highest setting (orange and yellow).
This is the sound of a low pass filter sweeping over a drum loop:
Fig. 12. Reason’s high pass filter rolls off frequencies below 20 Hz at its lowest setting (allowing frequencies in the yellow and orange areas to pass) to 4 kHz at its highest (the orange area only). The slope attenuates 18 dB per octave.
This is the sound of a high pass filter sweeping over a drum loop:
EQ types: shelving
If you’ve ever had a stereo or car radio with bass and treble controls, shelving is another type of EQ you’re already familiar with. (If all you’ve ever had is an iPod, we’ll talk about the type of EQ you use the most — graphic EQ — in a bit.) Shelving EQ is usually used in the common “treble and bass” configuration. It’s also used at the upper and lower ends of EQ systems that have more than just two bands. Like low pass and high pass filters, shelving EQ works on the “from here to the end” model: Everything below a particular point (in the case of a bass control) is affected, and everything above a particular point (in the case of a treble control) is affected. The difference is that shelving EQ boosts or cuts the levels of the affected frequencies by an amount that you specify; it doesn’t just block them entirely, which is what a pass filter does.
Shelving EQ is the perfect tool to use when a track has energy in one of the extreme registers that you want to emphasize (boost) or reduce (cut), but you don’t want to target the specific frequencies or eliminate them entirely. It lets you keep the overall level of the track at the level you want compared with your other tracks in the mix, while giving you a quick way to distinguish or disguise the track. Some useful applications include:
- Percussion tracks often have energy in the extreme low and extreme high frequency areas; shelving EQ can easily bring that energy to the fore of your mix or cut it to make room for the sound of another track
- Synth bass parts make or break a dance track; a little boost with shelving EQ can quickly transform a dull track to a booty-shaker
- Adding a high shelf to a drum kit and then cutting by a few dB gives the kit a muffled, alternative mood
Let’s take a look at and give a listen to the ways that shelving EQ works.
Fig. 13. The two-band EQ on the Reason’s 14:2 Mixer device is a classic example of treble and bass shelving EQ. Turning a knob clockwise boosts the affected frequencies, and turning a knob counterclockwise cuts the affected frequencies.
Fig. 14. This diagram gives you an idea of how shelving EQ differs from simple pass filter EQ, using the specs of the 14:2 EQ. The middle of the yellow area is the part of the track that is unaffected by the EQ; you’ll still hear the frequencies in this area at the same level, even if you boost or cut the frequencies in the shelving areas. The blue area shows the frequencies affected by the bass control: Below 80 Hz, you can cut or boost the frequencies by up to 24 dB (dark blue lines). The orange section shows the area affected by the treble control: Above 12 kHz, you can boost or cut by 24 dB (red lines).
On this drum loop, first we’ll cut the high shelving EQ, then we’ll boost it. Next we’ll cut the low shelving EQ, then we’ll boost it:
Fig. 15. Reason also gives you shelving EQ on each mixer channel. This EQ has a bit more control than that on the 14:2 mixer, as it allows you to specify the frequency above or below which the track is affected. For the high shelf, you can adjust the cutoff frequency from 1.5 kHz to 22 kHz. The bass shelf cutoff can move from 40 Hz to 600 Hz. In both cases, you can cut or boost by 20 dB.
EQ type: parametric EQ
So far, our EQ tools have been like blunt instruments: Effective when fairly wide frequency bands are your target. But what if you have just a few overtones that you need to reduce in a single track, or an octave in the mid range that you’d love to be able to boost just a tad across the entire mix?
That’s where parametric EQ comes in. Parametric EQ usually divides the frequency spectrum up into bands. Some EQs have just two bands, like the PEQ-2 Two-Band Parametric EQ device. Some EQs have as many as seven or eight bands. Usually, three or four bands will give you all the power you need.
For each band of parametric EQ, you can select three parameters: the center frequency of the band, the amount of cut or boost, and the bandwidth, which is often referred to as the “Q”. Adjustments in Q typically are expressed on a scale that approaches 0.0 and 3.0 at either end, with 0.0 being the widest bandwidth and the gentlest slope and 3.0 being the narrowest bandwidth with the steepest slope. Let’s see what shapes parametric bands can get.
Fig. 16. This is Reason’s channel strip EQ, highlighting the parametric EQ.
Fig. 17. This is the curve of a band of parametric EQ, centered on 1 kHz and given its highest boost with the widest Q setting. Looks like a nice, graceful old volcano.
Fig. 18. This curve has the same center frequency and boost amount as the previous diagram, but with the narrowest bandwidth, or Q setting. This is what’s known as a spike.
The illustrations above are meant to show you the full extent of the power that parametric EQ can offer. Normally, there’s no need to use all of it. In the vast majority of cases, just a tiny bit of EQ adjustment will have a huge effect on your music, no matter what type of EQ you use.
But for now, let’s use this power to explore the concept of overtones a little further. Sometimes it’s hard to believe that every musical sound you hear — with the exception of the simplest waveforms — consists of a fundamental frequency and lots of overtones, each at their own frequency. An instrument playing a single note is just a single note, isn’t it?
Let’s take a listen to a single pitch played on a piano. We’ll hit each note hard and let it decay for just a moment. While we play, we’ll take one of those narrow bands of parametric EQ, the one with the spiky shape, and we’ll boost it and sweep it up and down the frequency spectrum. Listen closely, and you’ll hear several overtones of this single pitch become amplified.
Sweeping a piano with an EQ spike:
Sweeping the EQ spike across the single piano pitch makes it sound more like an arpeggio than a single pitch. All those “other notes” are the overtones. Even in one single note, you have a multitude of frequencies that give the sound its character. In fact, when you hear a person describe a sound as something that “cuts through in the mix,” they’re usually referring to a sound that’s rich in overtones, more than a sound that’s simply louder than all the rest of the sounds in the mix. Usually it refers to a live performance P.A. mix, but the idea applies to what we’re doing here, too.
Let’s experiment with cutting overtones, using the same EQ curve, but reversed. Listen to the same repeated piano note as we sweep the parametric EQ across the frequency spectrum.
Sweeping a narrow parametric cut on a piano sound:
As we swept the EQ cut up and down, you could hear the piano tone change radically. Sometimes it sounded hollow, sometimes as though it had too much bass. But there were other times when it still sounded good, even though we were reducing the sound of particular overtones.
The key to creating a great sounding mix is to know how to cut or boost overtones on each track to make each instrument have its own sonic space, so that its overtones don’t interfere with other tracks, and other overtones don’t interfere with it. Using parametric EQ to cut particular overtones in one track to make room for the overtones in another track is the one of the most effective ways to do it.
Let’s create a mix right now, using only the types of EQ we’ve discussed so far. We won’t touch any faders, nor will we pan any tracks. Just by manipulating overtones by cutting with EQ — we’re not even going to boost anything — we’ll turn a muddy mix into a clear one.
Create a mix using only EQ
Here’s an excerpt of the raw, unmixed multitrack of a blues session, recorded using Reason. Check out the raw tracks.
Raw blues tracks excerpt:
Usually when you think of Muddy and the blues, you think of the great Muddy Waters. This session, on the other hand, was the other kind of muddy. The kick drum, bass, and rhythm guitar are overwhelmed by the thick sound of the horn section. The solo organ and tenor sax parts don’t overlap, but their tone sure does. Sounds like we have our work cut out for us.
Where do you start? Since the bass and kick drum are the least audible, you might be tempted to bring up their levels with a shelving EQ. Don’t bring any levels up! We’re on a mission to cut and carve out space for each part.
In fact, just to get you comfortable with using EQ, we’re not only going to cut, we’re going to cut a lot, using as much as the full 20 dB of range! Normally, just a couple dB will do the trick. But let’s go hog wild, just to prove that EQ can’t hurt you.
Let’s start with the mud. The horn section sounds great, but they’re not the most important thing going on. We’ll solo them along with the part they seem to be obliterating the most: the guitar.
Guitar and horn section soloed:
Yes, that’s muddy. The guitar itself is boomy as heck, too. Using the low mid frequency parametric EQ band on each track’s channel strip, let’s employ our narrow-Q parametric EQ cut-sweeping trick. The guitar sounds much better with a big cut at 260 Hz, and the horns open up for the guitar when they’re cut at 410 Hz for the tenor sax, 480 Hz for the alto sax, and 1 kHz for the trumpet. The saxes benefit from a slightly wider Q. Here’s how just these four sound, with EQ cuts.
Guitar and horn section with EQ:
Fig. 19. These are the low mid frequency EQ settings for the guitar, tenor sax, alto sax, and trumpet, from left to right, skipping over the darker channel. The amount of cut is radical, but it’s for educational purposes. Plus, it sounds pretty good in the mix, as you’ll hear.
Overall, these four tracks sound a bit thin, especially the horns. Keep in mind we’re overdoing the EQ amount. But the horn section still sounds like a horn section, the guitar is audible finally, and you can hear the bass guitar now, too. The organ and solo tenor sax are way out in front now, which gives us some room to maneuver with them.
Guitar and horn section with all tracks:
Since we want the tenor sax to be the more prominent of the two, let’s solo the two tracks and then carve some space for them.
Tenor sax solo and organ fills:
It’s obvious these two instruments are stepping on each other’s sonic toes, so to speak. Let’s apply our narrow-Q parametric EQ cut-sweeping trick to the organ, and cut it at 412 Hz or so, which sounds the best after the up-and-down sweep. That opens up the tenor sax, and yet the organ is still very audible. Since the tenor sax has such a rich sound, let’s apply a low shelf filter cut to it, right at 155 Hz. Now the balance sounds right between the two solo instruments, and there’s probably more room for the rest of the mix, too.
Solo sax and fill organ, EQed:
Fig. 20. These are the low mid frequency EQ settings for the organ fills on the right, and the low filter shelving settings for the tenor sax on the left.
Now the mix is really opening up. You can hear the electric bass and drums much more clearly, and though the tenor sax is definitely front-and-center, you still hear the horns, guitar, and organ.
Solo sax and fill organ with all tracks:
Since the drummer is using the ride cymbal, there is a lot of high-mid energy in the drum part. Let’s see if we can cut some of that and still have the nice “ping” of the stick on the cymbal — using the narrow-Q parametric EQ cut-sweeping trick, of course. First, the drums, soloed:
Raw drums, soloed:
After sweeping up and down on the soloed drums, we didn’t really find a setting that removed the “hiss” but preserved the “ping” of the ride cymbal. So we applied a low pass filter instead, with a cutoff setting of 4.9 kHz. This got us the sound we were after.
Drums with low pass filter:
Fig. 21. A low pass filter worked best on these drums; this is the setting.
Now let’s listen to the mix with all of our EQ work in place:
All tracks with EQ.
All the tracks have their own sonic space. Every instrument is audible, and no track is obscuring any other track. It’s a pretty good rough mix, and we didn’t even make use of panning, reverb, dynamics, or level settings! That means that the distance from this mix to a final mix is much closer than it would have been if we’d started with adjusting levels — and we still have all the fader headroom to work with!
To be sure, we applied EQ in far too generous amounts. But that was just to show you how powerful yet easy it is to make a muddy mix into a transparent one.
It’s also good to note that Reason’s mixer is perfect for this kind of EQing: The controls are all visible at all times, and you can see which track might still have room for a center frequency adjustment.
But even more impressive is what you can’t see, but you can certainly hear: Reason’s EQ section sounds fabulous. It’s very smooth, and very musical.
EQ types: graphic EQ
If you’re an iTunes user, you know all about graphic EQ. It’s similar to parametric EQ in that you target specific bands of frequencies to cut or boost. But the frequencies are hard-wired. Some graphic EQs have as few as five or six bands, while others divide the audio spectrum up into 30 or more bands.
Fig. 22. This is the graphic EQ in Apple iTunes. Graphic EQ is great for applying an overall setting to an entire concert, or to a lot of songs in a particular genre.
Graphic EQ is most often used in live P.A. work, where the room is subject to resonance points that are constant. A graphic EQ lets the engineer identify those points quickly and then cut those risky frequencies from the entire mix for the rest of the night.
In mixing, you need EQ that you can tailor more to the music itself, which is why parametric EQ is what you find on great mixing boards. Like the one in Reason.
EQ types: mastering EQ
Once your mix has been perfected by your skillful use of EQ, dynamics, reverb, panning, and level settings, it’s ready to be mastered — which is another way of saying it’s ready for the final polish. Usually, parametric EQ is used at the mastering stage, but it’s used very sparingly. It’s great for bringing out some of the middle frequencies that perhaps seem to ultimately get lost in the mix after all your work.
Reason has a special EQ device for mastering: the MClass Equalizer. It’s automatically patched in to the insert effects in the master channel, along with the other MClass devices: Stereo Imager, Compressor, and Maximizer. Used together with subtle settings, the MClass devices can add a magical, professional polish to your mix.
Ernie Rideout is currently writing Reason Power!, scheduled to be published in early 2010 by Cengage Learning. He grapples with words and music in the San Francisco Bay Area.