Advanced Drum Mixing: Time-Aligning Your Drum Tracks for Better Phase Coherence

When it comes to recording live musicians, few instruments are as misunderstood or mysterious as the drums.

The drum kit presents a complexity that just doesn’t exist when recording other instruments, simply because of the sheer number of components that make up even the most basic drum setup, as well as its wide frequency response and abundance of transients.

Even when you’re using a single microphone to record a full drum kit, there are still several drums and cymbals (not to mention the room) that need consideration and attention.

To try and maintain control over the balance of the drum kit, we recording engineers will often use multiple mics to allow us to manipulate the level and tone of each component while, hopefully, maintaining some sense of realism or believability.

But as you add more microphones, you also increase the likelihood of destructive interactions between the microphones, putting your whole sound at risk. These destructive interactions between mics are caused by phase cancellation.

The best way to combat phase cancellation is by placing your mics carefully to begin with. However, if you’re left with significant phase issues when it’s time to mix, a little bit of time-alignment can dramatically improve the clarity, coherence and impact of your drum sound. As you become more and more adept at this technique, you may even find that well-recorded drums can benefit from a little time spent experimenting with alignment.

To hear the sound of drums before and after time-alignment, listen below:

In just a few moments, we’ll look at a clear-step-by-step process for time aligning drums that works every time. But first, a little bit of background:

Phase Cancellation: The Arch Enemy of the Perfect Drum Sound

When you combine two signals of equal amplitude and frequency that are “inphase” with one another, they will reinforce each other, and your signal will get louder as a result. This is known as “constructive interference.

On the other hand, if you combine those same signals but they are 180˚ out of phase with each other they will cancel each other out completely. This is destructive interference or phase cancellation.

This basic concept is easy to understand by combining two sine waves of the same frequency and timing, or using two identical audio tracks. Simply open your DAW, take any audio file, and duplicate it so you have two copies. Then, invert the polarity on one track and bring up both faders to the same level. The combined output will be -∞, or rather, nothing.

This same idea is still at play when you are using multiple microphones on a full-frequency source like a drum kit, though becomes slightly more complicated. Not only are the signals more complex, but also there are multiple arrival times to deal with.

Multiple arrival times at each microphone can make phase coherence for multi-tracked a much more complicated affair.

Each mic on the kit will “hear” each drum and cymbal at a different level, at a different time, from a different angle, with a different frequency response. This creates multiple instances of constructive and destructive interference simultaneously.

The results can be unpredictable: Some frequencies will cancel completely while others will be slightly reduced; still others will be reinforced, and so on.

To make things even more confusing, the signals are not simply “in” or “out” of phase with each other: They can be out of phase by anywhere from 1˚-359˚, and by different amounts at different frequencies!

This is not a problem that can be easily fixed with just a polarity switch, either. Without the right techniques to deal with it, it can be a chaotic mess.

This kind of chaos is more common than you might expect, and less experienced recording engineers will often make matters worse by working on the sound of each mic individually without regard to how it will interact with the other mics as they are added to the drum mix.

For example, a new mixer might bring up the kick drum, make it sound good by itself, then solo the snare drum, make it sound good by itself, and so on and so on, down the line until each mic is “perfect”.

However, once you bring every fader back up and listen to the drum mix, there is shock and awe. (And not the good kind). This approach usually results in tracks that don’t fit together well, and when combined, the sound is nothing like the sum of its parts.

Drum tracks like this often suffer from uneven bass response, peaky midrange, a lack of high frequency detail, a hollow or non-existent center image in stereo, and a general absence of size and impact. These are not the adjectives you want to use when describing your work to prospective clients!

More experienced engineers will usually take a more comprehensive approach to working on a drum sound. They will pay attention to the overall balance, and show awareness of how the addition of each mic affects the balance and tone of the drum mix. They may still work on the sound of each mic, but only after the cumulative effects of all of the mics have been accounted for and dealt with through proper mic placement.

Better Phase Coherence = Less Work!

The most obvious benefit to ensuring that you have good phase coherence at the tracking stage is that your drum tracks end up needing less EQ and other processing to have impact and size, because everything is operating as a whole. It’s one drum sound, not just a bunch of drums combined (like you may get when using samples). The sound is more coherent, clear and focused, and yes, bigger.

Pages: First | 1 | 2 | 3 | ... | Next → | Last | Single Page

  • Eric Seaberg

    Without listening to the demos, I totally disagree. How many YEARS of classic hits have been recorded without the ability to shift digital files back-n-forth to match phase? Of course you know by changing the timing, you’re also changing ‘stereo’ imaging information?

    I’ve been in the biz for 45-years, spending lots of time in LA, Hollywood and Burbank when labels had BUDGETS to record albums… if the technology had been available at the time, I can guarantee it never would’ve been used because the PERFORMANCE of the song always outweighed the TECHNOLOGICAL CORRECTNESS of the song… and that is what’s wrong with the music industry today. Everyone has forgotten about the performance and FEEL of the MUSIC.

    As perfect as you think you can get the ‘digital imprint’ of a song, you’ll never be able to make it better if the performance sucks…

  • Mike Major

    I too have been in the music biz for a long time too (30 years) and started in the analog world where no such luxury was available. I always got it right with mic choice and positioning. There’s truly no substitute for that, I agree.

    However, since I am often mixing tracks that people didn’t record in a studio with a veteran engineer I am often faced with a bunch of tracks that don’t fit together no matter what I do. This process helps correct, or at least minimize the deficiencies that occur when you have people recording drums who may be novices or who may not as careful or methodical as someone with experience.

    And of course this changes the stereo imaging information-I talk about that and what to pay attention to when you’re making these adjustments. The fact of the matter is that adding more mics without regard to how they fit together will also change the stereo image (which happens too often). By being methodical about going through them (as you would when mixing any tracks) but knowing that you’re not stuck with the sometimes bad, destructive decisions related to the time-domain that people make when they put mics in bad places is nothing but beneficial to the mix.

    Nowhere in the article do I suggest that this is a cure-all for a bad performance, bad musicians or a crummy song. I said nothing about the feel or the music. It’s merely a technique that I developed and use every day that makes the tracks fit together so the drum mix has more coherence. That’s all.

    I also said nothing about technological correctness, either; I said that this makes the drums sound better. The technical descriptions are merely to give the reader some context about how and why to make the decisions instead of taking a stab in the dark.

    I also said nothing about making things perfect. Just better.

    And even when there’s a budget, sometimes the tracks don’t fit together.

  • George Piazza

    With the advent of smaller recording rooms, and subsequent decrease in budgets and big studios, for many, these techniques offers a way to improve drums sounds that are often unobtainable with the average modern recording setup. Though I agree somewhat with Eric Seaberg’s assessment, the fact is it is impractical to expect even seasoned recording / mixing engineers to get a great phase coherent multi mic drum recording in this age of limited budgets and time constraints. The other factor is that many older beloved drum recordings were done with minimal mic techniques, with excellent mics, preamps & drum-kits in finely tuned rooms. And those recordings are all over the map in terms of spectral content, stereo spread & definition.
    For most working engineers, timing & phase adjustments are a fact of life in the contemporary recording / mixing world. And engineers have been checking polarity since it was an available option during recording and / or mixing. That is the very least a good mixing engineer should do when pulling together the drum sound on a recording.
    Recording & mixing is an evolving art; drum replacement was already prevalent in the early 90’s. Today’s mixes generally demand more power, focus and stereo spread than recording of 20 + years ago. Though we could stand to back off on hyper-compression, bringing multi miked drum tracks into better coherence (without dumping all spatial cues) is a part of contemporary ‘good sound’.
    Besides the Little Labs IBP, I would also recommend the Voxengo PHA979, a linear phase rotator & timing adjuster with a 1/3 octave FFT phase coherence spectrum display. It rotates all the frequencies by the same amount, unlike the minimum phase IBP.

  • Vic Steffens

    I’ve been in the business 48 years….I WIN!!!!!!!!. No seriously, I started as a live drummer, became a session drummer, transitioned to engineering and producing. Wa last years Producer f the Year in New England. Not that any of this bullshit matter…it just proves I’m old.

    I have recorded a long list of great drummers successfully, from Steve Gadd to Mickey Curry, to Joe Franco..I got a list. With all do respect the concept of shifting drums around has always seemed ridiculous to me. I will grant you that in a mix situation you MIGHT be faced with something done so poorly you have no choice. Sure, any port in a storm. But of course in todays world you just use your drum replacement ax of choice. Once you go there…its a whole new ballgame….but I don’t think that is what we are talking about.

    So why do I think this is nuts? First off….you are supposed to hear the kit as one instrument, not nine or ten. Reflections and time discrepancies are part of the sound of the kit. Those discrepancies generally make the kit sound fuller and bigger. Yes you can shift something and maybe make it sound better with another part of the kit, but that relationship you are changing is probably doing something not so good to another part of the kit.

    So sure, you wanna play, have fun, but when you start writing articles that encourage it IMHOP a bunch of people that read this stuff will suddenly think that whatever crap you record can be magically saved in post. Someone will make a
    drum saver” plug or something. Its not all that hard to do it right. Put your time into that.

    Sorry to act like an old curmudgeon. Wait….I AM an old curmudgeon……

  • Mike Major

    My Grandfather was 99 years old. He clearly wins.

    If you listen to the examples above you can hear what I’m talking about in the article. The first example is not as dramatic but in the second one the snare is clearly fuller and more present and the overall sound has more impact. There are no other adjustments to the mix in the two examples; faders and panning are unchanged. In sample two it’s not a subtle difference. Seriously.

    And the article pretty clearly states that this is something to do if things are not right when it comes time to mix.

    Quote from the article:

    The best way to combat phase cancellation is by placing your mics carefully to begin with. However, if you’re left with significant phase issues when it’s time to mix, a little bit of time-alignment can dramatically improve the clarity, coherence and impact of your drum sound.

    I am not in any way implying that you can “magically save” anything with this technique. The idea is that you can improve the relationship between the mics. Sometimes significantly…other times, not so much. This is as close as you can get to moving mics around while you’re miking the drums. But since I’m talking about mixing tracks that have already been recorded, that is not an option.

    When I did live sound years ago, we eventually used a digital crossover/drive system which, of course, gave us the ability to delay the PA system to match to stage sound and backline. Every time we did it, it sounded better and took less EQ because you weren’t fighting the differing arrival times and the inherent phase and EQ problems that they caused. I’ve simply applied the same concept to getting a bunch of drum mics to work together better.

    This technique does make the drums sound more like one instrument since all the mics play together better. It also helps you to use less EQ and other crap to make the drums sound bigger and fuller, which is always better in my book. And the improvement may keep some from using samples at all. I rarely, if ever use samples because a) they’re evil, b) no matter how good they get, they just match the dynamics of the real thing, c) I am more interested in using something unique and original that came from the band, not a sample library.

    The only way to avoid destructive phase problems when recording drums is to use one mic only, but most people don’t record that way and most contemporary drummers don’t want to hear their drums that way either.

    The whole idea is to make the drums sound better, and usually, more like they did in the room.

  • Dietz Mix

    Great advice! That’s one of the few examples where the WWW’s signal is indeed louder than its noise. 😉 Like Mike points out, time-alignment and phase-correction can make a good multi-mic recording great, and they can make mediocre ones usable (at least).

    … one small objection has to be raised, though: Absolute phase _does_ matter. Especially in case of a really LF-rich kick drum a positive first half-wave will literally push the air out of the speakers (assuming that they are wired correctly). Flipping the phase of the kick (read: first half-wave negative) will “suck” the air away from the listener, thus reducing that sensation of a good “thump into the stomach” quite a bit.

    What’s more, the phase relations between the kick drum and the bass (… the instrument, not the frequency range) will matter a lot, too. Keeping an eye on them will it make easier to create powerful, punchy mixes without lots of processing in the actual sense of the word.

    PS: The all-important disclaimer: I’m in the mixing business since 30 years, and my grandfather would be around 120 by now. ;-D … and sorry for any strange neologisms – English is not my first language.

  • Mike Major

    Thanks Dietz Mix, I appreciate the kind words about the article.

    You know, you say that English is not your first language, while using the word “neologisms” in the same sentence! I think you’re doing fine…

    With regard to the absolute phase thing: I am speaking from my personal experience more than anything empirical, and simply put, I leave things where they sound best rather than seeing which direction the waveform or the speaker happens to be moving. I get your point though. A friend of mine used to wonder about this too: if the kick drum pushes air out of the front of the drum then the mic diaphragm will move inward (relative to itself, of course) which should then cause the speaker to move outward. It’s a strange relationship when considering what’s absolute phase, no?

    I hear that some turtles live a very long life; sometimes 150 years. Imagine what they must know about recording!