Working With The Spectrum

The best way to get better with using EQs is by listening to how EQ adjustments alter a track or sound. I’ve heard some people suggest creating a steep bell-curve (see the image below) and boosting the gain, sweeping the frequency spectrum back and forth to listen for problem areas. I think that this method is actually more useful for simply getting used to what parts of the spectrum sound like. Creating a frequency sweep allows you to hear what things sound like at 1000hz versus 500hz. For easy reference, check out my blog post that discusses the frequency spectrum. Or just check out the guide below.

40-100hz: This is the thud of your sub-bass and kick drum. Boosting this too much and you’ll clip your meters and make a mess of your low end. Keep the sub-bass levels too low and your track will feel like it’s missing something when played over a sound system.

100-200hz: This is your standard bass range. You can boost this area for additional snare punch, or to increase the beefiness of your bass lines. I find that often the kick drum has a ‘knock’ in this range.

200-500hz: The higher end of bass frequencies. Lower synth lines can be situated in this range, and boosting too much can create jumbled sounds. Too little in this area leads to a weak sound. Boosting the bass harmonics in this area can help them stand out on sound systems with poor low end capabilities.

500-1000hz: The range where a lot of things are usually happening. Boosting too much can lead to nasal or awkward sounds. Too little though will make a weak mix. This area is often where lead synths sit in tracks.

1-2khz: An area where the body of clap and snare sounds sit. EQing this area can be tricky because you want your percussion and drum elements to stand out but not interfere with lead synth sounds. But high-passing elements in this area can lead to thin drum sounds. This is also where the lower ranges of hi-hats and cymbals tend to sit. Don’t boost too much in this range though because this is where human hearing is most sensitive.

2-5khz: This is where vocals often sit. They go lower in the spectrum too, but this is where the clarity comes from. Adding a boost will give an edge to your vocals and synth lines, but cutting too much will result in somewhat of a muted sound.

5-10khz: This area contains the crispness of snares, the brightness of hi-hats, clarity of synths and vocals, and the top-most part of bass sounds. Boosting will help these elements stand out more, but too much can lead to an unwanted scratchiness. Likewise, cutting too much will lead to a muted or dull mix. You also have the sibilance of singers, that ‘s’sound. Too much will be harsh on the ears, but too little makes the vocals harder to understand.

10+khz: This is the part of the frequency spectrum that provides breathiness, air, and a certain brilliance or glitter. When you think your mix sounds dull or like something is missing, but you can’t place it, it is often this range of the spectrum. On the opposite end, too much presence sounds hissy and artificial in a bad way

Understanding The Spectrum

The secret to working with EQs and understanding the frequency range is to be able to visualize it. Having a spectrum analyzer is tremendously beneficial while producing. Not only does it show you where different elements are loudest, but you can then compare different channels to spot where overlap is occurring. The frequency spectrum begins at 20 hertz on the far left and continues until 20khz at the far right.

This is considered the typical range of what humans can hear, and by seeing it as a spectrum, you can plop your different sounds into their own space along that spectrum. Hi-hats and cymbals are on the right, bass lines and kick drums are far left. If you can squeeze your sounds in nice and smooth so they don’t overlap much, you’ll be well on your way to a fine-sounding mix.

For the science-minded, I like to imagine a room with different gas elements. The thicker elements are hovering just above the floor, the thinner ones are light and airy, and hang by the ceiling. Your bass elements are the thicker, deeper sounds. They rest near the floor. The hi-hats and cymbals are breathy and light, and they are at the top of the room. The point I’m trying to make is that if you can visualize where your sounds are along the spectrum, fitting them together will make your life a lot easier.

EQ Modes

Depending on your choice of equalizer, there are probably a few different modes available. This section covers the most common ones.

Regular EQ: This kind of EQ is most likely the default setting on whichever equalizer you prefer to use. It’s stereo output and making adjustments doesn’t affect the stereo image.

Left/Right EQ: This splits the stereo image into left and right channels, allowing you to make adjustments on each individual channel. This can be used for creative purposes, like when trying to cut a frequency range in the left side but boost it in the right.

Mid-Side EQ: A mid-side EQ is fundamental for advanced productions. Rather than splitting the stereo image between left and right, a mid-side splits the image between mono and stereo. So all your sounds down the ‘mid’dle are adjusted using the mid EQ, and all your sounds on the ‘side’ are adjusted using the side EQ. This comes in handy when you want to keep your low end in mono, or when you want to duplicate a synth line, but apply a bunch of effects and then take out the mono parts so as to add a bit of atmosphere.

Minimal vs. Linear Phasing: In the days of analog gear, making cuts or boosts with equalizers introduced phase-shifts. This occurred because when you cut or boost a part of a sound, you introduce a very slight latency, or delay, to the signal. So your output would be different parts of the same signal occurring at very slightly different times, thus coloring the sound and shifting the phase. If the sound coloration was desirable, then there were no problems whatsoever. The issue is that it was typically an unwanted effect, and so EQ manufacturers took strides to reduce the phase-shifting that would occur, hence the term ‘minimum-phase’. A linear phase EQ allows you to introduce latency to all frequencies, so that your output is the same across the board and your phase does not shift. Unfortunately, this can cause issues in your low end for reasons that are beyond the scope of this book. Typically, linear phase EQs are preferable for mastering engineers, but otherwise aren’t worth the effort. Stick to a regular EQ and you’ll be fine. You probably won’t even notice any phase shifting that might occur.

EQ Types

As discussed above, the primary equalizer styles seen in action today are parametric and graphic EQs. But there are other types as well.

● Graphic: Graphic EQs have a simple interface. A handful of sliders at specified frequencies that you can adjust up or down to boost or cut the frequency band.
● Parametric: A visual representation of an EQ, usually partnered with a spectrum analysis. Frequencies, Q-values, and gain are all entirely adjustable.
● Program: Program EQs are less common, but you’ve probably seen one at some point. It’s really just the technical term for the bass and treble knobs on your home stereo system. A program EQ is designed so that you only have control over the boost or cut amount, not the frequency band or width.
● Shelving: A shelving EQ is a simple equalizer with a high- or low-shelf filter.

Cutoff Filter: When one thinks of a filter used as an effect, rather than an EQ, this is what comes to mind. It’s typically a high-pass or low-pass filter with a high Q-value.

Filter Types

To start, there are five basic filter types. High- and low-pass, shelving, peak, notch, and band-pass filters.

Low-pass and high-pass filters: These filters reduce the gain on any frequencies below or above the cut-off point. There are different ‘slope’settings that measure how drastic the cut or gain reduction is. Depending on the EQ, it is typical to see a range from 6dB/Octave all the way up to 48dB/Octave. As explained earlier, this means that for every octave, six (or whatever the number is) additional decibels of gain reduction are applied. Also worth noting:The actual cutoff point of a filter is defined as a 3dB reduction, so the reduction actually starts a bit before whatever your cutoff frequency is.

Low- and high-shelf filters: Similar to a low- or high-pass filter, a shelf filter applies gain reduction beyond a certain frequency. But a shelf filter can also be used to boost frequencies too. The primary difference between cutoff and shelving filters is that the reduction is equal at all points on a shelf filter, not gradually increased. So, if you apply a low-shelf filter at 200hz with a reduction of 6dB, there will only be a 6dB reduction at 100hz too, rather than a 12dB reduction from a low-pass filter.

Peak (or bell curve) filters: A peak filter allows one to isolate problem frequencies. This filter type is the kind that lets you make narrow cuts or wide boosts. You can change the width by adjusting the Q-value.

Notch filters: Notch filters make very steep, narrow cuts in a frequency range. Using this filter type is great for surgical cuts, but exercise caution because it can result in a very odd, fake sound. Really, any major cuts or giant boosts can result in your instrument sounding wonky if not done right.

Band-pass filters: Finally, the band-pass filter. This filter type is a combination of both high-pass and low-pass filters, allowing only a specific range of frequencies through, while attenuating the entire range on either ends of the band-pass.

Using EQs: How

This section covers all the knobs and dials, and what each part’s function is. If you’re familiar with this stuff already, it may be best to skip to the next section. To begin, an EQ is comprised of multiple filters. There are different kinds of filters, but we’ll get to that in a moment. Each filter has at least three facets that determine where, how intense, and how widespread the adjustment is. They are:

Center Frequency: The center frequency is the point at which you will be making a cut or boost. This tells you where in the frequency spectrum the filter will act. For example, a small cut at 500hz tells you that the center frequency is 500hz.

Q-Value (Resonance): The Q-value measures the width of the cut or boost. Is it a very steep cut? This would be a high Q-value. The broader the curve, the lower the Q-value.

Gain: Lastly, each filter needs to know the amount of gain to be applied. A cut will have a negative value, such as -3dB, whereas a boost will have a positive value.

Slope: With certain filter types, there is also the slope value. The slope value determines how much of a gain reduction will be applied as you progress through the spectrum. The slope value is measured by the number of decibels reduced per octave. For example, a filter with a cutoff slope of 12dB/Octave will reduce the gain by 12dB every octave below the center frequency. Since octaves are half or double the last frequency, depending on the direction, a cutoff at 400hz with a 12dB/Octave slope would result in a gain reduction of 12dB at 200hz, a 24dB reduction at 100hz, and a 36dB reduction at 50hz.

With those details out of the way, let’s discuss the types of filters found in an equalizer.

Using EQs: When

Once you know why an EQ is pertinent to dance music production, you next need to figure out when it’s wise to use one. This is where a lot of producers and artists get stuck. You can find tons of charts online, articles on what frequencies cause problems, and guru rules to follow. None of this matters though if you aren’t certain why you are EQing in the first place.

The best answer you’ll ever get for when to use an EQ is this: Use your ears. I know it sounds like a poor excuse, but it’s really not. If you can’t rely on your ears, then you need to get better at training yourself. If you need training, try this exercise:

1. Load up two channels in your DAW. Drop a synthesizer onto each channel.
2. Create a synth melody or loop in one channel, and play white noise in the other channel.

3. Tell me how good it sounds. The secret is that it won’t sound good at all. It’ll be messy and muddy and sound horrible. That’s an example of your ears telling you something isn’t right. Another exercise is to create a song with five different lead synth lines playing at once, all in the same octave. It won’t sound good, and not just because it’s five different melodies. If, on the other hand, you do trust your ears to make reliable decisions, then you’re ready to start learning when to use an equalizer. Every song or loop you create, you should ask yourself what doesn’t sound amazing. Does your instrument sound like crap? Is your bass line thin and weak? Do your hi-hats have too much hiss? Are the cymbals drowning out your snare? If you can nail the diagnosis, EQing will be easy.

Learning to listen to your music will get you far in the production universe. Obviously, when it comes to EQing an instrument, the best scenario is when you don’t need to use an EQ at all. Ideally, your elements sound fine, and they sit well together in the frequency spectrum. Unfortunately, that won’t always be the case, and you must learn to determine what you need to correct or change. One way to improve this skill is by sitting down and listening to every song you make. Pinpoint exactly what you’re hearing.

Do certain elements need to be quieter or louder? Where does each instrument sit in the frequency range? Does everything appear close or far from you? Writing down your results will teach you a lot. Additionally, before you begin EQing, figure out what your focal point is in each song. Having a focal point will help you decide which instruments need to be prominent in each part of the frequency range. If you struggle with finding a focal point, try the following exercise:

1. Number each channel or instrument in your song in order of importance.
2. The top channel will be dominant (what you want your listener to focus on), and other channels will support it.
3. Examine where your primary instrument sits in the frequency range, and then look to see what else is occupying the same space.
4. Try EQing the non-prominent instruments to make room for your lead. Alternatively, it might be that you only need to reduce the volume on certain instruments rather than EQ them.

Once you can train your ears to understand what sounds good and what sounds muddy or messy, you can then go about creating music, merely EQing as necessary when certain sounds need help sitting together.

Using EQs: Why

There are two primary uses for an equalizer when it comes to electronic music production. The first is what you probably think of initially – troubleshooting and fixing sounds. Whenever you have overlapping or interfering instruments, using an EQ will help make your elements play well together. If an audio sample is too harsh, or there are clicks and hisses, you can use an EQ to remove those problems. This is the troubleshooting role of an EQ.

The other role is to add color, character, or generally just use an equalizer as an effect rather than a problem-solving tool. Vintage analog EQs (and their plugin emulations) provide color and sound shaping in many interesting ways. Or, you might want to use an EQ in all the ways people say not to, just to see what happens. This is the creative, colorizing, character-adding role of an EQ.

And of course there are different EQ models and plugins that are designed specifically for one of the above roles. Some are meant to add character. Others are prime for fixing issues. Through experience you will learn when to use each of your EQs.

What Is An EQ’s Role?

Here’s the simplest definition I can come up with: An equalizer basically takes an incoming signal and adds or removes gain at certain frequency bands, and then outputs the adjusted signal. Imagine a straight, flat line. That’s your pre-EQ signal. Now, bend the line, creating hills and troughs in it, and that’s what an EQ does. In the two images below, the first one shows a flat signal, the second shows an adjusted signal.

At its simplest, an equalizer is simply a complex volume control. You can raise or lower the volume on different areas of the frequency spectrum. So, why would anybody use an equalizer? Great question! Keep reading

Graphic and Parametric EQs

While we’re on this topic, allow me to briefly explain the difference between a graphic and parametric EQ.

These days, graphic equalizers are more suited for live recording environments where one needs to make adjustments quickly. A graphic EQ has many bands with gain sliders at different frequencies. See the image below for a modern graphic EQ.

The idea is to adjust the sliders so as to make a smooth curve, which then roughly resembles the actual curve of your frequency shaping.

A parametric EQ is more common, especially with digital workstations. The equalizer bands are easier to manipulate and adjust, and working visually tends to produce results easier as well. An example of a parametric EQ would be Ableton’s EQ-8. Parametric equalizers often have a combination of high-pass and low-pass filters, shelf filters, and peak (or bell curve) filters. Some might even have notch filters as well.