How to Disappear Completely: My Year Without EQ

Chris Mitchell, FOH Engineer for Umphrey’s McGee

This past weekend marks the one year anniversary of removing the last few bands of corrective parametric equalization from the input channels of my FoH mix for Umphreys McGee. I spent two full years working out how to give up corrective EQ. Now I feel ready to explain how I did it.

It started with what I consider the core of sound for a rock band: kick drum, snare, and hi-hats. These three parts of the drum kit establish the core tempo and rhythm that the sound of the band builds from. The kick drum should be ever present, solid but natural, with the lowest fundamental tone combined with plenty of beater contact. The snare drum should have a texture and complexity – an aggressive snap, followed by a thick shell decay and a crispy wire rattle from the snares. The hi-hats are delicate and expressive. The combination of a spring-loaded foot pedal and face-to-face mirror-image cymbals allows for a multitude of complex tonal options. Their decay envelope varies with pedal tension, from a short tick to a loose, splashy wash that works well in rock ‘n roll.

Add to this foundation a quartet of tom-tom drums, rich and deep, with the clean snap of uncoated heads. The low richness of the skins is contrasted by a brassy chorus of cymbals, each with its own distinct voice. I wanted my drums to sound like the drums on my favorite albums from the 70s & 80s. I wanted Aja drums. I wanted Tom Sawyer drums. Or my ultimate comparison, the drum tones Sylvia Massy got on Ænima – dynamic and impactful with rich overtones and beautiful decays. I asked myself what the difference was between those drum sounds and my drum sounds. What qualities do I need to hone? The answer starts at the destination, not the source.

Music is enjoyed in our minds. The emotions we seek are created when our brains translate airborne pressure vibrations into synaptic excitation and then compare those to previous experience. Our brains do an extremely good job discerning the difference between sounds produced by the original source and sounds emanating from a reproduction of the original source. I want the original sound of the band to be accurately represented in my mix. I do not want the audience to be distracted by the mix. It is a sound engineer’s goal to become invisible and facilitate an emotional connection between artist and audience, regardless of production level.

For UM, this means that the drum kit in my mix should sound exactly like Kris Myers’ perfectly tuned Pearl Reference drum kit. I want the drums to sound like the listener is slightly above and in front of the drum kit, but at the volume I choose. To get this, I designed a way to combine multiple sources that would result in the drum mix I want.

Every mix engineer recalls the advice handed down from on high: It’s better to move the microphone than move the EQ knob. If the mic doesn’t sound good on the source, swap it for another that sounds better. But who has time to reposition the microphone when soundcheck is in 20 minutes? Not to mention that most mic lockers are stocked to be homogenous and interchangeable, with the same predictable options. So what are we looking for when we move or swap the mics? Remember, I’m striving for truth in reproduction. Assuming the drummer (or tech) can tune a drum kit to sound great, then I should just have to make it louder, without changing the tone of any of the drums or cymbals. Choose a microphone that has flat frequency response, both on and off axis.

These improvements produced the drum sound I wanted. The mix became organic and dynamic. When Kris plays softer or louder his dynamics and tones are accurately reproduced. Every stick impact and shell decay sounds faithful and sharp.

It’s relatively easy to twist and shape tones and levels to make a drum kit sound like almost anything. It’s much harder to remove any evidence that someone is behind the curtain. I’m reminded of my Navy Electronics course and a class about amplifier theory. The discussion was about a hypothetically perfect circuit described as “a wire with gain.” A device designed to pass a signal with no loss, addition or distortion – a signal that does not lose any integrity when amplified. That’s how I think about my drum mix now: it should be just like standing in front of the kit, at whatever volume I desire. The accurate translation of a small sound to a large experience, like the PA isn’t there and the band is playing just for the fan.

Of course, a band is more than just a drum kit, and I had to make this approach work for every other instrument in the mix.

I knew that Earthworks microphones could help me solve that problem. Known for their ruler flat response and extremely accurate translation of fast transients, Earthworks SR series of Cardioid and Hypercardioid microphones are a perfect choice for an accurate, truthful mix. One of the first mic changes I made was switching the drum overheads from AKG C414B ULSII’s to Earthworks SR30’s and from a traditional separated pair of stands to an XY coincidental pair suspended directly over the snare drum mic. I was immediately impressed by how much more of the overheads I could use in my mix.

Soon I muted the hi-hat, ride cymbal and snare bottom mics. Didn’t need them. The Earthworks SR30’s were providing an awesome stereo image of the top of the drum kit. Plenty of hi-hats and cymbals combined with the pop and snap of the snare and tom-toms. The “close” mics on the snare and tom-toms were chosen to reinforce the rich low mid overtones produced by the head and shells. I finally settled on a pair of Earthworks DP30/C’s on the rack toms and Audix D6’s on the floor toms, with another DP30/C on the snare top and a Telefunken M80SH on the aux snare.

With the advent of digital mixing consoles, many new features have entered the audio engineer’s toolbox. One of my favorites is input channel delay. If you’ve ever aligned multiple speakers in a system, you realize the importance of time alignment between components. Without properly adjusted delay times, cancellations and summations create problems that cannot be solved with equalization. The same can be true for input sources.

Now that I was running my overheads at a high level in my mix (fader @+3db usually) I thought about delaying my snare and tom-tom mics to allow the waveforms to time up with the waveforms picked up by the overheads.

I’ll support this with a little math. Imagine the stick hitting the snare drum. The snare top mic is 2″ away from the drum head and the sound wave takes 0.15ms to reach it. The overheads (XY) are an additional 44″ away from the snare mic, taking 3.26ms more for the wave to reach the microphones. By adding 3.26ms of delay to the snare top mic, I effectively align the waveforms in time, reducing the problems caused by phase cancellations and summations. I tried this and liked the results so much that I built a spreadsheet to help me calculate the delay times needed for the tom-tom and aux snare mics too. Not a huge change, but an incremental improvement towards the perfect drum kit sound. And to make sure I can get the same results every show, I use a heavy Atlas mic stand with positions marked in silver Sharpie.

The kick drum was pretty straightforward with a few exceptions. I use the common dual mic technique, with a Shure Beta91A attached to a small pillow inside the kick drum. I use an Earthworks Kickpad to shape the tone and face the mic away from the beater head. That position better balances the tone of the shell and beater. I high pass this mic at 160Hz, using it only for attack and shell tones. For a long time I used an Audix D6 in the front head hole for the bottom end. The D6 has been a long-time favorite. But I wanted a lower fundamental drum tone without resorting to EQ. After a few experiments, I built my own Subkick-like transducer with a resonant frequency (Fs) of 38Hz, dropping smoothly to around 80Hz, gone by 160Hz. Combine that with the Beta91A, which has 0.59ms of delay to compensate for the distance between the transducers, and the result is a thick, solid tone that never gets swallowed by the Bass guitar or Moog keyboards.

The timbre and tone of the drum kit were starting to come into focus. But I couldn’t ignore dynamics. One of the hard parts of mixing a rock band is getting the dynamic range of the different parts of the band to sit in the mix with each other. The drum kit is a very dynamic instrument, ranging from soft snare and cymbal rolls all the way up to bombastic cacophony, maybe 40-50db between quiet and loud. But the guitar amps may only range 25-30db. Bass and keys even less. Compressors help us match these up, but sometimes at the expense of tone and transient reproduction. It’s very easy to compress every channel and wrestle the drum kit into submission, but the old adage applies, “If your only tool is a hammer, every problem looks like a nail”.

I decided to trust the dynamic capabilities of our drummer and use no channel compression. I set up parallel subgroup compression to thicken the drum sounds and raise the RMS level, reducing overall dynamic range but allowing the excitement of uncompressed drum channels to pop in the mix. This technique improved when I switched to the Earthworks mics. The transient response of the capsule accurately captures the very quickly changing waveform, impossible with the much heavier capsule used in dynamic mics like the Shure SM57 (my least favorite microphone for drums or percussion).