Time alignment / Phase alignment. Picks and Audio examples.

tarnationsauce2

Welcome to the jungle.
DRUMS - Time alignment / Phase corrction. Pics and Audio examples!

Time, distance, and sound pressure… all very well defined variables, but also very hard to control variables. Luckily we have a DAW to correct some things that would take hours upon hours of mic placement in the old days. And well make the impossible possible, like making the snare mic coincident with an OH mic 6feet away.

In this example I used:
Kick - AKG D112
Snare – SM57
Under snare – CAD M179 set to cardioid pad engaged
Over Heads - Spaced pair 2x Stellar CM-1 set to cardioid (Apex 460 variant)
Toms - CAD M177 pad engaged

The audio is completely dry, no compression, no mixing, no panning, no EQ. Completely stock.

The ONLY differences between the audio files are time alignment, and phase alignment. This should help demonstrate how these variables affect frequency response, power, and imaging.

Here are some pictures of what an un-aligned vs. aligned waves look like. Phase, well that’s just a button on my DAW so it doesn’t show it graphically. But I’m sure you already know what it looks like when you invert phase.

Go ahead and load the files in your audio editing software to “see” the level difference if you can’t hear it. But I bet you’ll hear it. ;)

The stock, untouched mix:

The mix with only time alignment on all tracks, even toms:

The mix with time alignment, as well as phase inverted on under snare and kick:


The picture below is split into 4 sections. I didn’t post pics of the toms, but it’s the same deal.
Section 1 Shows the stock time alignment.
Section 2 Shows the snare / bottom snare aligned to the overheads the overheads always should not be moved. Notice the under snare is inverted?
Section 3 Shows stock kick alignment. There is a lot less kick power in the overheads, that’s why it is so small. But alignment still helps. You can also see that the phase is pretty close to inverted here too.
Section 4 Shows the kick aligned to the overheads. Again, do not move the overheads, only move the kick track.

You'll also notice the overheads themselves have the snare a bit sooner on one side than the other, that can even be corrected. In this case it's a bout an 7 sample shift. Probably not even worth moving. This is best to align by better mic placement in relation to the snare.
 

Attachments

  • time_phase_pic.gif
    time_phase_pic.gif
    34.7 KB · Views: 282
Last edited:
I'll bet Dragon could free up some server space with all the lost threads that are hiding there somewhere......
 
I was actually just thinking about making a thread about this. I was wondering how exactly you should move things. All the close miked stuff to the overheads, or the overheads to the close miked things?


Actually here's another thing I'm curious about. If you have a room mic, what do you do with that?
 
Last edited:
Time alignment (and thus phase coherencey) can make a MASSIVE difference to your sound. it can be a bit of a painful process, but well worth it. Consider making a really transient noise just before the section of interest is recorded (for example, a stick click or something similar. This will allow you to roughly time align the kit.


For a closer, more exact time alignment go through each piece of the kit hitting it once or twice. Then you need to time align each close mic to the OHs (because they'll be different distances from the mic to the OHs, they're phase shift will be different).

However, if you're going to start getting really anal about things, you'll soon find that this doesn't work: the hihats will hit one OH sooner than the other, as will any piece of kit which isn't exactly in the center of the OHs..
 
I just took a song I recorded and went back and lined up the tracks. I used the OH's as the starting point and brought the snare up to that and then the kick to match the OH's. You really have to zoom the crap out of you tracks before you can see how off they are. I think doing this really tightens things up. The snare now seems more focused and the kick has more punch.

While I think this is a quick and easy thing to do and to my ears it sounds better, I have been reading up on this for a week now and some guys swear by it and some guys say it's the devil.

For room mics I think it's better to leave them alone. Lining them up would defeat the purpose. At least that is what I have read. I have never tried to use a room mic. Never had a good enough sounding room and never had enough channels.
 
Is time-alignment and all these adjustments something that only became neccassarry with digital recording? Or is this something people in the know have been doing since analog???
 
RAMI said:
Is time-alignment and all these adjustments something that only became neccassarry with digital recording? Or is this something people in the know have been doing since analog???


These delays have always been present (they quite simply reflect the physical distance seperating mics). It's just become much easier to do it.
 
MessianicDreams said:
These delays have always been present (they quite simply reflect the physical distance seperating mics). It's just become much easier to do it.
I see. Yeah, I didn't think a signal would take a different amount of time to hit a mic just because it's going into a digital recorder. :eek: It just seems that I hear more about time alignment lately. Probably because, like you said, it's easier to see and fix now.

So did guys like George martin or whoever from the 60's and 70's somehow take care of it? Or was there a reason it was less of an issue?

I'm a curious bastard....If only I could retain the answers to all my questions... :( :mad:
 
I think back in the days before home studios there were guys that knew how to place mics in the right place and knew what to listen for.
This time alignment thing is just another tool that a guy like me can use to compensate for my less than stellar audio engineering skills. ;)
 
RAMI said:
So did guys like George martin or whoever from the 60's and 70's somehow take care of it? Or was there a reason it was less of an issue?
A big part of why it was less of an issue then is because massive multi-miking of drum kits was a much rarer phenomenon. That really didn't be come super popular until the late '70s and 80s, along with the total number of available tracks to record on.

G.
 
EdWonbass said:
I think back in the days before home studios there were guys that knew how to place mics in the right place and knew what to listen for.
This time alignment thing is just another tool that a guy like me can use to compensate for my less than stellar audio engineering skills. ;)

The way I see it is that unless you time align the tracks you MUST - and can at least to some degree - hide the effects of comb filtering by clever mic placing or/and EQing.

The main thing to understand is that there's just two variables that affect the strength and nature of the resulting comb filter effect. The time difference defines how closely spaced the dips in the frequency response are. For example 1ms delay causes a dip at every 1000Hz, 5ms delay every 200Hz and 10ms delay every 100Hz and so forth.

In the end it's the relative strength of the two signals at a dip frequency that determines how deep that particular dip is and when the signals are of equal strength the dip is the deepest but when the signals differ more than 10dB the deptht of the dip is less than 1dB.

Let's take a simple scenario where you want to bring up the snare in the mix a few dBs. Without the knowlege of the facts above when you raise the fader of the close miced snare track as soon as you hear the snare getting louder you'll start to hear the unwanted side effects caused by comb filtering and it gets worse and worse until the signals are of equal strengt and then starts to dimish gradually.

Now how can you save this situation? Basically the only thing to do is to try to make the signals' strength as unequal as possible at dip frequencies by using eq to cut the other signal or boost the other or by choosing a snare mic with the boost in the right freguency bands.

OTOH If you do the time align thing first then there would be no need to try to hide the side effects just raise the fader and the snare will get louder. In the end what matters the most is that the end result of summing the signals is much more predictable when there's no comb filtering to take into account.
 
Chris. said:
All the close miked stuff to the overheads, or the overheads to the close miked things?
the overheads are always the "stationary" part. Think of them as the "master" clock. the reason is, all the mics have a different time/distance offset to them. So you have to establish what of all the waves will be your wave to align to.
Chris. said:
Actually here's another thing I'm curious about. If you have a room mic, what do you do with that?
The whole idea behind a room mic is the delay from the overheads, and the room reverberations it'll capture since it's not close mic'd. So as previously stated, that would partially defeat the purpose of a room mic. However if you want to get rid of any amount of delay and comb filtering, align it too. You'll still get the natural room verb.
 
Back
Top