Time Alignment & the Drum Kit

If you mix on a digital console, this post is REALLY important.  Unless you’ve been mixing under a rock, I’m sure you’ve heard about time aligning your drum mics.  I’ve been reading about others’ results with this for a long time but have never been faithful to time align every time I mix because most of the samples I’ve heard of before and after have been less than convincing.  My prejudice, and maybe yours as well if you’ve heard similar samples, is that time alignment on the console using channel delays is a 2% mix improvement step, not a 15% or larger mix thing.  Because of our intense pace at Kensington, it has been hard to have the discipline to time align every time.

When I was mixing on a PM1D every week, I made an effort each time to time align by ear and just slide back the snare, hat, and tom mics back to the overheads by anywhere from 2-6 milliseconds.  The difference was noticeable but still what I would consider a minor “glue” thing – certainly not earth-shattering.  When I moved to the Venue, I kind of set the time alignment stuff aside for a few months in order to focus on more core tasks and efficiencies.  A few months ago, I began to revisit time alignment and this time, for some reason, the results really were a game changer.  I’m not sure exactly what to attribute that to, but I can share some tips on the overall approach I’ve settled on that seems to give the best results.  If you haven’t tried this, I would suggest you should and see if it makes as big a difference for you as it did for me.

My goal in time aligning is so that the tone of a drum (let’s use the snare as an example) will sound nearly the same when listening to the close mic as when the overheads are added in.  What I found a few months ago in the run up to Easter was that it was bothering me to get a snare sound I liked with the close mic and by the time the overheads were where I wanted them that tone was destroyed.  Pull the overheads out and the snare tone would go back to where I wanted it.  What I found was that I was hearing a combination of overheads with too low of a high-pass filter setting and comb filtering between the close mic and overheads.

I should preface that there are two different philosophies for overheads on a drum kit.  The first is to place them strictly as cymbal mics.  For some music styles or drummers, this is necessary in order to get the expected mix results.  The other option is to use them as an overall kit mic that just so happens to have cymbals as a focus point.  In this scenario, the high-pass filter will likely be set lower than it would for cymbal micing applications – probably 100-200 hz instead of 500 hz plus for cymbals.  Another important part of the kit mic approach is using mics in a X-Y pattern over the kit.  I go back and forth between a Rode NT4 and Shure KSM141s on the Shure stereo adaptor bar.

The first step to the results I want is to use the overheads as kit mics and still set the high-pass around 250 hz.  This allows the overheads to still be the glue for the kit but the close mics to provide the “in your face” body of the sound.

Next, each mic is delayed back to the overheads.  I’ve found the easiest way for me to do this is to have the drummer give me a quarter notes on each drum and record that into PT.  (If you don’t have PT, instead of this technique, just go drum by drum and have the player give quarter notes while you dial each mic in)  Once in PT, I zoom in closely on those hits and measure the difference in time between each of the close mics and the overhead.  The results are usually somewhere between 40-160 samples.  (You could also do this with milliseconds but because PT & Venue can go to samples resolution, that’s what I use)  When you have it right, its the first time in my life that there is a negligible difference between the tone of the close mic and once you put the overheads in.

Next time I’ll post some samples that will hopefully convince you.  This topic has been written about all over the web so if I’ve piqued your interest, start digging into it and feel free to ask questions here.

The great PA experiment

Ever since we purchased our last rig at Kensington a few years ago for the opening of our West campus, our vendor has suggested we should bring it in to the Troy main campus and see how it would perform in a larger space.  This past week I finally made that happen.

If you’ve been following me for a while, you know the current PA rig at the main campus is tired at best.  We went through an exhaustive design process last year to plan for an expansion and upgrade to the campus but due to the current economic realities of our area, those dreams have been on hold indefinitely since late last year.  On top of this, we’ve been especially struggling lately with questions of volume and volume quality as it relates to everything we do in the room.  This test provided a unique opportunity to evaluate how a more ideal speaker solution would effect the audio experience.

The test PA in question was a L’Acoustics Kiva/Kilo/SB118 rig.  6 Kivas & 2 Kilos per side, 2 SB118s on the ground per side.

I hung the PA on Sunday afternoon as a part of the stage turn from the weekend to midweek.  On Tuesday morning we connected everything up and sent signal through for the first time.  One important part of this test was to keep our current rig completely intact so we could simply A/B between the two different systems.

The long and short of it, as to be expected, is that the Kiva/Kilo system was like listening to a set of nearfield monitors in the room, while the KF650s sounded like there were a couple mattresses stacked up in front of them keeping you from hearing any direct sound and instead deflecting all the energy everywhere but at your ears.  The clarity and musicality was something I’ve never heard in the room before.  The room became FAR less of a factor because it didn’t feel like you were listening to the room – you were truly hearing what was coming from the speakers.  Subtle EQ translated, TL Space verbs sounded amazing on vocals, and everything just sounded like it should BEFORE you touched channel EQ.  The low end was tight and defined.  Kick and bass each had a place in the mix and someone said you could actually hear the string noise on the bass rather than just hearing tones that you assume are coming from a guitar.

The tests on Tuesday afternoon confirmed everything I suspected about the importance of the correct PA solution for the room and we’d never see the kind of progress leadership is looking for with volume without it.

We decided to go with the system for the midweek service and the results were pretty solid.  The entire night averaged 5 dB less than normal (mid-80′s to low 90′s rather than topping out at 95-97 dBA) but it didn’t feel like we compromised anything artistically.  With our 650s, there is a point you have to run at in order to have some degree of clarity and get over all of the interaction with the room.  A line array in the room all but removes that variable.  At the end of the service, we had an extended response time with broken down acoustic/vocal and the teacher trading back and forth.  This moment was the most beautiful thing I’ve ever heard in this space.  I pulled everything WAY back so the music was in the high 60s-low 70s dBA and it felt perfect for the mood, but the vocal had every bit the same clarity, presence, and body that it had when we were 20-30 dB hotter earlier in the night.  Amazing!

Learnings from this test is that, in light of economic realities, we need to consider how a smaller and lower cost rig would fit into our room permanently.  The Kiva/Kilo combo wasn’t my favorite box, per se; I felt like it was a bit harsh and bright.  However, it was certainly enough box to get what we need from it on a week to week basis.  The 4 SB118s were a bit light on the low end, though, but that’s to be expected.  I also felt like I wished they extended down a bit lower then they did. We’ll have to experiment with the correct type of sub and quantity to get this right.

The next test is a set of L’Acoustics ARC’s that will be in the room for a few services.  Even though I know the pattern of these boxes is not a good fit long-term for our room, I’m interested to see what we think after living with something different for more than a single service.

The best part is that this test set some further ideas in motion.  A bit premature to discuss right now, but there will be more thoughts later…