Parametric EQ 101 or: I thought Q was a Star Trek character?!

Jared Taylor —  September 11, 2014 — Leave a comment


We’ve already done a 3-part series on EQ, and we covered the critical listening skills required to use it, instrument by instrument, in our series “Taking It Apart”. These posts are great learning material, and I highly recommend looking back at them.

Our locations are getting new digital sound consoles. This is exciting for a number of reasons, one of which is EQ. Our old consoles were analog and had semi-parametric EQ, which means they had some of the functionality of a parametric EQ, but were missing a few features. In our case, the high and low frequency bands were fixed shelf filters. The high mid and low-mid bands were sweepable filters, but there were no width or Q adjustments.

Okay stop — if that last sentence confused the heck out of you, you really need to go back and read at least the overview post from the original EQ series.

Great. Now, that you know what I mean by band filters, shelving filters and everybody’s favourite omniscient alien life form (Q), you can read on. We’ll take a quick look at the fully parametric EQ controls you’ll find on a console like the Allen & Heath Qu16.

Boost and Cut

Boost and Cut

A simple definition for EQ is a volume control for a specific frequency. This definition serves us fairly well with semi-parametric EQ.

A semi-parametric EQ has a gain boost/cut for each band, and some bands allow you to change the frequency you’re adjusting with a sweep control.

The tricky part is you can never adjust only one frequency. EQ bands are centred on a particular frequency, but they always affect a range of nearby frequencies.

Sweep from high to low frequencies

Sweep from high to low frequencies

So a better definition for EQ would be a volume control for a specific range of frequencies.

To make things even more complicated, the width gets narrower as the boost increases, resulting in the shape of a slope. A fully-parametric EQ has a control for the shape of this slope. It can be called width which is measured in octaves, or Q which is a ratio of the centre frequency to the width of the band. You may see the value measured in fractions or decimals of octaves, or you may see Q values ranging from 0.1 to 100. However your device chooses to display the values, if you turn up the knob, the band gets narrower. Turn it down and the band gets wider. This is counter-intuitive for most people–it still is for me and I’ve been doing this for a long time!

Narrow and Wide Q, or width, filters

Narrow and Wide Q, or width, filters

If you’re still getting used to boosts, cuts and sweeping, I suggest you set your width controls to the middle and leave them alone. You need to grow your ears for a while before you take this one on! But if you’re more experienced, learning to tweak the width can make a big difference for your mix.


How wide and narrow bands sound

Wide bands tend to shape the tone in a more natural way. When a frequency is boosted or cut with a wide bandwidth, lots of neighbouring frequencies come along for the ride, resulting in a gradual slope from the centre of the boost to the edges that are left flat. It’s a smoother transition. Older EQs almost always used wide bands, so plugins and presets that use words like “Classic” or “Vintage” are usually employing, among other things, wide filters.

Narrow bands tend to sound more drastic and are often used to cut problem frequencies. Problem frequencies don’t sound natural to begin with, so a narrow width cut is appropriate to eliminate them without affecting those happy neighbouring frequencies you’d like to keep in the mix. I rarely, if ever, boost a very narrow frequency (unless I’m making a screen shot for this blog). The result usually sounds unnatural.


A word about why: the musical side

I love working with consoles that measure width in octaves because it reminds me of how closely EQ is linked to the musical scale.

Think about it: in concert pitch, the A4 note occurs at 440 Hz, or cycles per second. This is the same 440 Hz we can dial in with an EQ because frequency, pitch and note are all different words the same thing. So we can boost or cut a range of frequencies in and around a note like A4 with an EQ. In fact, the narrowest width setting on a console like the Qu16 is 1/9th of an octave, which is darn close to bandwidth of a single note. If you made a big cut at 1/9th octave at 440 Hz, you would effectively reduce the volume of the A4 note on the instrument you were EQing! Of course, notes on an instrument contain other frequencies, called overtones or harmonics, but the fundamental note is the frequency that defines the pitch for our ears. On a bass guitar, those fundamentals come through very loud in the mix and it’s not hard to accidentally boost or cut a note or two in the bass guitar’s range.

This changes the audience’s perception of a performance, making certain notes louder than others. Let me underscore that this is not a good thing! Such drastic filtering should be avoided.

On the other end of the width knob, our friend the Qu16 ranges up to 1.5 octaves. A talented and trained singer may be able to sing three octaves. An electric guitar can access up to four, and really only the keyboard uses more notes than that (although in a pop/rock band they probably shouldn’t). A wide band filter is still centred on a particular pitch, but it can’t single out individual notes. Instead, a default width of 3/4 of an octave–the default on many devices–is a smooth slope that gradually introduces the change across a range of frequencies. This is much more useful for tone-shaping, where the goal is to balance the harmonics and overtones of the instrument on the whole. If you make a cut at, say, 2 kHz (’cause you can never go wrong cutting at 2 kHz) you can smoothly de-emphasize that annoying range of frequencies without noticeably cutting out the B6 note, whose fundamental is 2 kHz.

Here’s the bottom line: if you need to eliminate one particular troublesome sound, go ahead and narrow down the width and have at it. But if you’re trying to shape the tonality of your source, you’ll find more natural results with wider bands.

Shelving filters

Shelving filters

One more thing a fully parametric EQ may offer, and almost certainly does if you’re working with a digital console, is the ability to select the type of filter for the high and low bands. Most analog desks have a shelving filter on the high and low frequency bands. Shelving filters are quite useful in these ranges, but many EQs allow you to switch them to band (bell) filters. Either filter type may be useful in different situations (I still tend to prefer the shelves) but it’s important to note that many consoles default to bell filters, which is probably a big change if you’re used to an analog desk.

Band, or Bell filters

Band, or Bell filters

Hopefully this helps you understand the basics of parametric EQ. I feel compelled to mention again that tone is always best adjusted at the source. If you can tweak mic choice or position, adjust amp or a keyboard patch or change your guitar strings to get closer to the right sound, start there!

Happy mixing!

Jared Taylor

Posts Twitter Facebook Google+

No Comments

Be the first to start the conversation.

Leave a Reply

Text formatting is available via select HTML. <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>